SPECTROSCOPY AND SPECTRAL ANALYSIS

News

Forgot your password?

Enter your e-mail address to receive an e-mail with your account information.

Journal Online

Archive

Links

	CAST

	CNCOS

	CNKI

	WanfangDATA

	CHEMSOC

	sinospectroscopy

	CPS

Quick Search

Adv Search

2025 Vol. 45, No. 05 Published: 2025-05-01

	1201	Research Progress for Online Monitoring and Calibration Technology on Microbial Aerosol With Fluorescence Method
		HU Xing-zhi^{1, 2}, LIU Xiao-meng^2*, ZHANG Sheng-zi², XIANG Jun¹, WANG Hong-jun²
		DOI: 10.3964/j.issn.1000-0593(2025)05-1201-08
		Microbial aerosol surveillance is essential to ensure national security and human health. The online monitoring method based on fluorescence detection technology stands out in microbial aerosol monitoring due to its advantages of fast response, strong portability, and high sensitivity. This method accurately identifies individual bioaerosol particles and monitors real-time aerosol content in the environment. Its working principle is to use ultraviolet lasers to induce the intrinsic fluorescence of bioparticles and then detect the intrinsic fluorescence signals to achieve the identification of bioparticles. Although fluorescence online monitoring equipment has significant advantages in monitoring the concentration and particle size of biological particles in ambient air, the technological development in this field in China is still in its infancy. In the past 20 years, significant progress has been made in laser-induced online monitoring of microbial aerosols. These advances are mainly reflected in the optimization of excitation light cost, the subdivision of detection sorting channels, and the advancement of robotics and algorithm programming languages. However, the quantitative application of these instruments is limited by the lack of standard fluorescence calibration methods, which also increases the complexity of comparing different measurements. With the widespread application of online monitoring instruments, fluorescence aerosol detection instruments' performance evaluation and calibration technology have gradually attracted attention. The calibration research on fluorescence bioaerosol detection systems is still in its infancy, and no unified performance evaluation criteria have yet been formed. This paper briefly introduces the development of fluorescence online monitoring instruments, the research status of fluorescence-based single-particle online monitoring, and its calibration technology, which is summarized for the first time. It provides strong support for developing microbial aerosol detection technology and establishing a performance evaluation system for microbial aerosol detection instruments.
		2025 Vol. 45 (05): 1201-1208 [Abstract] ( 243 ) RICH HTML PDF (6196 KB) ( 200 )

	1209	Denoising of Second Harmonic Signals of the Absorption Spectrum Using the Frequency Decomposition Combined With the Savitzky-Golay Filtering
		TU Xing-hua, LUAN Xiao-chen, WANG Zhan
		DOI: 10.3964/j.issn.1000-0593(2025)05-1209-08
		To address the issue of noise interference in the detection of second harmonic signals from weak gas absorption spectral lines, this paper proposes a method combining frequency decomposition (FD) and Savitzky-Golay filtering (SG filtering), referred to as FD-SG filtering, to denoise noisy signals. Frequency decomposition is a mathematical method used to decompose complex signals, with the advantage of independently selecting the number of decompositions or solving non-recursively. SG filtering processes groups of data, improving data accuracy while maintaining signal trend and width. The proposed method first applies frequency decomposition to the noisy signal, followed by SG filtering for secondary denoising. Then, it reconstructs the denoised second harmonic signal from the filtered effective components. After frequency decomposition, the distribution of each component is analyzed, confirming that the effective signal mainly resides in the first component but still contains residual noise. By constructing an adjustment factor P to select the optimal SG filter length, the residual noise in the effective component is effectively removed. In experiments with CO₂ in the air using an absorption spectral line at 1 578.222 nm, significant noise was found in the second harmonic signal, especially with large spike noise at scan cycle junctions. Fitting curve subtraction and frequency decomposition showed limited noise reduction for large spikes. The combination of SG filtering and the relationship between the second harmonic and spike peak values demonstrated that SG filtering effectively reduces spike noise, achieving ideal denoising with the appropriate filter length. When detecting the second harmonic signal of exhaled CO₂, the turbulence caused by exhalation increased and jittered the signal amplitude. FD-SG filtering successfully suppressed the noise, yielding a smoother second harmonic signal. This verifies the algorithm's effectiveness in denoising second harmonic signals for practical weak gas detection, showing significant advantages in noise suppression and large spike noise reduction. This has positive implications for improving signal quality and system accuracy in weak gas detection.
		2025 Vol. 45 (05): 1209-1216 [Abstract] ( 56 ) RICH HTML PDF (6211 KB) ( 123 )

	1217	A Novel Strategy for Viral Detection in Acute Respiratory Infections: Combining SERS With Machine Learning
		JIANG Heng¹, LÜ Zi-wei¹, LI Yang², DONG Tuo^1*
		DOI: 10.3964/j.issn.1000-0593(2025)05-1217-08
		Rapid and accurate detection of common viruses causing acute respiratory infections (ARI) is crucial for public health prevention and control. Although traditional viral detection methods have partially met clinical needs, they often have limitations such as long detection times, high costs, or limited sensitivity. There is an urgent need for faster and more efficient detection methods. Surface-Enhanced Raman Spectroscopy (SERS) has become a research hotspot in viral detection due to its high sensitivity and specificity. This study aims to develop a novel and efficient detection strategy combining SERS technology with machine learning methods to achieve precise detection of Respiratory Syncytial Virus (RSV), Influenza A Virus (IFA), and Human Adenovirus (HAdV). The study employs citrate-stabilized silver nanoparticles (Ag@cit) and uses iodine ion incubation and calcium ion aggregation to prepare silver nanoparticles (Ag@ICNPs) as the SERS substrate. Ag@ICNPs have high-quality “hotspots” suitable for virus detection, enabling ultra-fast, highly sensitive, and label-free capture of characteristic fingerprint spectra of respiratory viruses. This study integrates machine learning methods with SERS technology to further improve detection efficiency and accuracy. By improving various machine learning algorithms, a virus classifier was successfully established, which can rapidly identify the three viruses within 3 minutes with a detection limit as low as 1.0×10² copies·mL^-1and an accuracy rate of 100%. Additionally, the concentration-dependent curves constructed based on the relationship between viral concentration and characteristic peak intensity showed good linearity (R² greater than 0.998), providing the possibility for quantifying virus content in samples. This is important for monitoring treatment efficacy and disease progression through changes in viral load in clinical settings. This study reveals the significant advantages of the combined application of “SERS@machine learning” in rapidly and precisely detecting respiratory viruses, offering a potentially valuable new approach for ARI clinical diagnosis. It is expected to become an important tool in future clinical diagnosis and public health prevention and control.
		2025 Vol. 45 (05): 1217-1224 [Abstract] ( 47 ) RICH HTML PDF (20632 KB) ( 147 )

	1225	Construction Method and Application of a Standardized Phytoplankton Spectral Library Based on Uniform Interpolation
		ZHANG Xiao-ling¹, WANG Si-qi¹, ZHAO Nan-jing^2*, YIN Gao-fang², DONG Ming^{2, 3}, WANG Xiang⁴, ZHANG Sheng-jun¹, CHEN Wei-jie¹
		DOI: 10.3964/j.issn.1000-0593(2025)05-1225-11
		Phytoplankton are crucial producers in marine ecosystems. The discrete three-dimensional fluorescence spectroscopy method addresses the high operational requirements of continuous three-dimensional fluorescence spectrophotometers, thus meeting the needs for online and in-situ phytoplankton monitoring. However, challenges persist due to the instability of living fluorescence and the similarity in pigment composition among different algal species. This reduces the spectral differentiation between algal groups in discrete three-dimensional fluorescence spectroscopy and limits its practical application. This study investigates the three-dimensional fluorescence spectral characteristics of Bacillariophyta and Dinophyta under various environmental conditions. It analyzes the limitations of the commonly used average concentration-normalized standardized spectral library construction methods (hereinafter referred to as the “average method”). Based on this, a standardized phytoplankton spectral library construction method is proposed using uniform interpolation (hereinafter referred to as the “interpolation method”). The optimal interpolation range and number of interpolations for the phytoplankton standardized spectral library were determined through experiments. The results show that an average coverage rate of 88% and an average correlation coefficient of 0.98 are achieved within the range of M±2S, which is identified as the optimal interpolation range. With 10 interpolations, measurement accuracy improves by nearly 13% compared to the average method, and the computation time is only 0.494 8 seconds, marking it as the optimal number of interpolations considering both accuracy and computational efficiency. Accordingly, a standardized spectral library for Bacillariophyta and Dinophyta was constructed based on uniform interpolation. Combined with non-negative least squares linear regression analysis, this method was used to interpret the three-dimensional fluorescence spectra of a series of algal samples with known concentrations, and the interpretation results were compared with those obtained using the average method. The results indicated that the interpolation method showed a correlation coefficient (k value) ranging from 0.828 to 1.149 and a determination coefficient (R²) ranging from 0.616 to 0.953 for five pure algal samples. The number of Bacillariophyta samples that failed to be identified decreased significantly from 21 (with the average method) to only 2, and the average absolute relative errors in the analysis of Bacillariophyta and Dinophyta were 36.9% and 30.7%, respectively, representing reductions of 20.6% and 19.0% compared to the average method. These results indicate the effectiveness of the uniform interpolation-based method in improving the accuracy of quantitative analysis of Bacillariophyta and Dinophyta. After validating the method's effectiveness in the laboratory, this uniform interpolation-based standardized spectral library construction method was extended to Cyanophyta, Chlorophyta, and Cryptophyta and applied to AFA during a cruise in the South China Sea. This research provides an effective technical method for the rapid and accurate classification and measurement of phytoplankton, offering strong scientific support for future marine ecological monitoring and environmental protection.
		2025 Vol. 45 (05): 1225-1235 [Abstract] ( 44 ) RICH HTML PDF (16178 KB) ( 22 )

	1236	Identification of Pinelliae Rhizoma Decoction Pieces by Hyperspectral Imaging Combined With Machine Learning
		LI Ruo-tong¹, HU Hui-qiang², CAO Shi-yu¹, LU Meng-yao¹, LIU Meng-ran¹, FU Jia-yue¹, MAO Xiao-bo², WANG Hai-bo^3, FU Ling^{1, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1236-07
		There are four kinds of decoction pieces for Pinelliae Rhizoma, namely Pinelliae Rhizoma, Pinelliae Rhizoma praeparatum cum alumine, Pinelliae Rhizoma praeparatum cum zingibere et alumine, and Pinelliae Rhizoma praeparatum, recorded in the current Chinese Pharmacopoeia (2020 edition). Due to their similar appearance and weak odor characteristics, it's easy to confuse their manufacturing management, market circulation, and clinical application. Because of the requirement for instruments, reagents, and intricate detection steps, exploring and establishing an accurate, rapid, and non-destructive detection method for Pinelliae Rhizoma decoction pieces is necessary.This paper used hyperspectral imaging combined with machine learning to identify the four kinds of Pinelliae Rhizom adecoction pieces. The principal component analysis (PCA) was utilized to extract features from the hyperspectral data, and support vector machine (SVM), logistic regression (LR), multi-layer perceptron (MLP), and random forest (RF) classification models were established based on the full-band data model. The accuracy of both training and test sets of four classification models was evaluated, along with an analysis ofthe principal component proportion of the four models under optimal performance. Additionally, the t-distributed stochastic neighbor embedding (t-SNE)visual dimensionality reduction analysiswas conducted on the hyperspectral data of the fourkinds of decoction pieces. The SVM, LR, MLP, and RF classification models based on PCA can achieve accurate identification for PinelliaeRhizoma decoction pieces. The accuracy of the test set is 80.76%, 96.45%, 96.59%, and 86.77%; in addition, the proportion of principal components is 60%, 80%, 70%, and 80%, respectively. The t-SNE analysis by dimensionality reduction showed that the components of Pinelliae Rhizoma praeparatum cum zingibere et alumine and Pinelliae Rhizoma praeparatum cum alumine were relatively close and partly changed compared with Pinelliae Rhizoma. However, the chemical composition of Pinelliae Rhizoma praeparatum changed greatly after processing, which was very different from the above three kinds of decoction pieces. These results also agree with the average spectral reflectance result. It's the first application of hyperspectral imaging combined with machine learning to develop a predictive model for different decoction pieces of Pinelliae Rhizoma. This approach -successfully identifies Pinelliae Rhizoma decoction pieces accurately, rapidly, and non-destructively, thereby providing a novel identification method and scientific foundation for these products' rational production, circulation, and clinical application.
		2025 Vol. 45 (05): 1236-1242 [Abstract] ( 56 ) RICH HTML PDF (5061 KB) ( 41 )

	1243	Analysis of Urine Sediment Samples Based on Microscopy Hyperspectral Imaging Technology
		DENG Ying-jiao¹, CHEN Jun², WANG Jian-sheng¹, HU Liu-ping³, ZHANG Qing¹, DU Yu-zhen³, WANG Yan¹, LI Qing-li^1*
		DOI: 10.3964/j.issn.1000-0593(2025)05-1243-08
		The analysis of urine components, called urine sediment, is paramount in clinical practice. By observing particles, cells, and crystals in urine sediment, doctors can obtain important information about the patient's urogenital health, which is crucial for diagnosing urogenital-related ailments. However, identifying urine sediment crystals heavily relies on medical professionals' manual observation under a microscope, which is time-consuming, subjective, and often inaccurate. Consequently, automated microscopic urine sediment image analysis using image analysis technology has gained significant attention. However, these methods rely solely on morphological information to classify crystal samples, making distinguishing between morphology-similar crystals difficult, resulting in low classification accuracy. Microscopic hyperspectral imaging technology integrates spatial and spectral information, revealing distinct spectral characteristics as different substances exhibit varying degrees of light absorption and scattering across different spectral bands. In this study, we introduced microscopic hyperspectral imaging technology to analyze urine sediment crystal samples and used a self-developed microscopic hyperspectral imaging system to acquire hyperspectral images. We collected microscopic hyperspectral image data of five urine sediment crystal sample types, including calcium oxalate, cystine, calcium phosphate, uric acid, and triple phosphate. Additionally, we trained four machine learning classifiers support vector machine (SVM), k-nearest neighbors (KNN), decision tree (DT), and neural network(NN) models on this dataset to classify the five types of urine sediment crystal samples. The classification accuracies of SVM, KNN, DT, and NN models for the five types of urine sediment crystals reached 0.959 8, 0.959 8, 0.982 9, and 0.991 7, respectively. Our research indicates that applying microscopic hyperspectral imaging technology to urine sediment sample analysis enables the acquisition of spatial information and facilitates the extraction of discriminative spectral features, thereby assisting physicians in microscopic examination and supporting the popularization of urine sediment microscopy techniques.
		2025 Vol. 45 (05): 1243-1250 [Abstract] ( 53 ) RICH HTML PDF (16650 KB) ( 18 )

	1251	Using DANN to Classify the Mango Varieties With NIR Spectroscopy
		LI Tong-le¹, CHEN Xiao², CHEN Xiao-jing¹, CHEN Xi¹, YUAN Lei-ming¹, SHI Wen¹, HUANG Guang-zao^1*
		DOI: 10.3964/j.issn.1000-0593(2025)05-1251-06
		Different cultivars of mangoes can not only represent different qualities but also produce different economic benefits. Traditional mango variety identification methods often rely more on the experience of practitioners and are time-consuming and laborious. Therefore, how to quickly classify mango cultivars is an emerging problem that needs to be solved. Near-infrared (NIR) spectroscopy technology is a fast and non-destructive approach. Users can often identify different mangoes by combining machine learning methods with near-infrared spectroscopy data. However, the NIR spectral information of the same variety of mangoes can vary due to variations in different instruments, seasons, and years. These differences result in a different distribution between the previously measured sample data (source domain) and the newly measured sample data (target domain). Consequently, the present classification model cannot correctly classify new mango samples. Domain adaptation methods can solve this problem of model inapplicability caused by different data distributions. This article focuses on the distribution differences of mango near-infrared spectroscopy data caused by factors such as working temperature and season. The domain adaptation methods can solve the problem of model unsuitability caused by different data distributions. This article used a deep domain adaptive neural network (DANN) model to solve this problem. The DANN model effectively achieves cross-domain sample classification models by aligning features between two domains through adversarial learning. This article compared DANN with unsupervised dynamic orthogonal projection (uDOP) and joint distribution adaptation (JDA), two traditional domain adaptation methods based on statistical learning. The experimental results of applying these three methods in this article showed that the DANN model could achieve a classification accuracy of 94% for the test set in the binary classification task of mango varieties. In the multi-classification task of mango varieties, the classification accuracy of the DANN model was over 10% higher than that of uDOP and JDA. The results indicated that the DANN model could effectively solve the problem of mango variety recognition caused by the different distribution of near-infrared spectral data between two fields.
		2025 Vol. 45 (05): 1251-1256 [Abstract] ( 49 ) RICH HTML PDF (8486 KB) ( 63 )

	1257	Output Demodulation Technology of Vernier Effect Sensor Based on Bi-LSTM Network
		ZENG Xin, GUO Mao-sen^*, ZHANG Xin, DING Hui, HU Hong-li
		DOI: 10.3964/j.issn.1000-0593(2025)05-1257-07
		To solve the problem of output demodulation of optical vernier sensors, this paper proposes a spectral data prediction technology based on a bidirectional long short-term memory (Bi-LSTM) network. By utilizing the predictive ability of the Bi-LSTM network for data sequences, a wide spectral range of spectral data prediction has been achieved, thus solving the technical problem of cursor sensors having difficulty achieving output demodulation due to the limited working spectral range of light sources or spectral scanning techniques.By using this method, as long as a limited wavelength range of sensor output spectra is collected, the trained Bi-LSTM model can accurately predict the envelope curve of the sensor output spectra over a wide wavelength range, greatly reducing the technical requirements for the working spectral range of the vernier sensor.The paper introduces the basic principle and implementation process of the Bi-LSTM network for output demodulation of vernier sensors. The experiment proves the accuracy of this method in predicting the spectral data output of vernier sensors. The maximum wavelength error between the predicted curve and the actual spectral envelope at the peak is about 0.02 nm, and the maximum amplitude error is only 0.058%. In addition, the paper also verified the generalization of the Bi-LSTM network for demodulating the output spectra of cursor sensors with different envelope periods. For the output spectra of cursor sensors with different envelope periods, the maximum prediction error was 0.02 nm, and the maximum root mean square error (RMSE) was 9.72×10^-5, proving that the trained Bi-LSTM network has accurate “predictability” and “tracking” for the output spectra of cursor sensors with different envelope periods. Comprehensive research papers have shown that in practice, as long as the wavelength range of the working light source can cover half of the spectral envelope period of the vernier sensor (which can be met in most cases), the Bi-LSTM network can accurately predict the output spectrum of the sensor over a wide spectral range, greatly reducing the requirement for the spectral range of the working light source (or other spectral scanning techniques) of the vernier sensor. The paper has solved the problem of the output demodulation spectrum range of the cursor sensor being too wide and has theoretical and practical significance for application.
		2025 Vol. 45 (05): 1257-1263 [Abstract] ( 39 ) RICH HTML PDF (3876 KB) ( 9 )

	1264	Surface Plasmon Resonance Sensor Based on Bimetallic Layer Au/Ag/MXene/WS₂/BP
		ZOU Dan-dan¹, BAI Yu-jie^{1, 2}, MA Si-fan¹, ZHU Si-qiang^1*, PAN Jian-bing³
		DOI: 10.3964/j.issn.1000-0593(2025)05-1264-06
		Salt content detection sensors are of great significance for the rapid and accurate detection of solution salt content, especially for monitoring the salt density of external insulation in power equipment at high altitude complex environments. However, the traditional method usually has the shortcomings of complex operation, long detection time, and low sensitivity, which makes it difficult to meet the needs of modern industrial production. To improve this problem, we propose a novel surface plasmon resonance (SPR) sensor that combines the advantages of bimetallic layers of gold (Au) and silver (Ag) with the unique properties of two-dimensional materials MXene, tungsten disulfide (WS₂) and black phosphorus (BP), such as high electrical conductivity, hydrophilicity, excellent thermal stability, and adsorption capacity. By optimizing the structure of each layer of the sensor and comparing the sensitivity of resonance spectra of different metal structures of the sensor, the results show that the bimetallic layer (Au/Ag)/MXene/WS₂/BP structure sensor has a sensitivity of up to 200°·RIU^-1. The sensor has a stronger detection ability, so it can accurately and quickly detect the salt content of the solution and provide strong support for environmental protection, resource utilization, and safety in production.
		2025 Vol. 45 (05): 1264-1269 [Abstract] ( 40 ) RICH HTML PDF (5553 KB) ( 10 )

	1270	Study of Fenfluramine Molecule Based on the Density Functional Theory
		GUAN Li-chang^{1, 2}, FENG Lei^{2, 3}, ZHAO Nan^1, JIANG Xue-mei^2
		DOI: 10.3964/j.issn.1000-0593(2025)05-1270-07
		Fenfluramine can inhibit appetite, so many merchants illegally added it to food for sale. After eating fenfluramine, it can cause liver dysfunction, valvular heart disease, primary pulmonary hypertension, and other diseases that seriously affect health. Therefore, studying fenfluramine molecules' structure, spectrum, and molecular excitation is of great practical significance. This work used the density functional theory (DFT) method with the B3LYP functional and 6-311++G(2d,2p) basis set for structural optimization. Furthermore, a series of studies were done on the structure, frontier orbits, Raman spectra, electrostatic potential, and UV spectra of the fenfluramine molecule. The basic structure information was obtained. The highest occupied orbital (HOMO) and the lowest unoccupied orbital (LUMO) were both alpha + beta orbits. The energy of HOMO, LUMO, and their energy gap was -6.25, -1.22 and 5.03 eV, respectively. There were two strong peaks at 756.5 and 1 003.5 cm^-1 in the experimental Raman spectrum. The strong peak at 756.5 cm^-1 was the symmetric deformation vibration of CF₃ and the asymmetric deformation vibration ofC═C on the benzene ring. The strong peak at 1 003. 5 cm^-1 was the symmetrical deformation vibration ofC═C on the benzene ring, the characteristic band of meta-disubstituted benzene. The linear fitting equation of experimental Raman spectra and calculated Raman spectra isy=0.988x+10.328, R²=0.999, showing good consistency. This work also discussed the electrostatic potential and excited state properties of fenfluramine. The fenfluramine molecule contained 17 electrostatic potential energy maximum points and 12 electrostatic potential energy minimum points. The surface area distribution of electrostatic potential energy in the range of -0.01～0.025 a.u. was relatively uniform. The UV spectra were mainly determined by the first, second, and third excited states, and the contribution of the second excited state was as high as 82.516%. The electron excitation characteristics were studied by using hole-electron analysis. It could be found that S0→S1 and S0→S2 were attributed to the n-pi^* charge-transfer excitation in the direction from the amino group to a benzene ring. S0→S3 was attributed to the superposition of the n-pi^* charge-transfer excitation in the direction from amino group to benzene ring, and the n-σ^* local excitation between ammonio to carbon chain nearby. These basic theoretical calculations provide a theoretical basis for not only the detection of illegally added fenfluramine in food but also the study of its derivatives.
		2025 Vol. 45 (05): 1270-1276 [Abstract] ( 40 ) RICH HTML PDF (10916 KB) ( 15 )

	1277	Preparation and Luminescence Properties of Eu³⁺ Doped ZnO/ZnS Phosphors
		ZHANG Lan^{1, 2}, WANG Xi-gui²
		DOI: 10.3964/j.issn.1000-0593(2025)05-1277-06
		Eu³⁺ doped sol-gel-precipitation methods prepared ZnO/ZnS red phosphors. XRD, FTIR, TEM, and EDS test methods characterized the structure of the red phosphors at different temperatures. The results showed that the red phosphors contain ZnO and ZnS, samples of the hexagonal phase structure. The luminescent properties and mechanism of the red phosphors were discussed by fluorescence spectrometer. The results showed that combining ZnO and ZnS can form new band structures and increase the composite probability, thus improving the luminescent properties of the red phosphors. Eu³⁺ was located in the non-inversion central lattice in the phosphors, which is dominated by the electric dipole transition⁵D₀-⁷F₂. The optimal Eu³⁺ doping amount of the phosphor is 0.1 (molar fraction), and the optimal annealing temperature is 800 ℃.
		2025 Vol. 45 (05): 1277-1282 [Abstract] ( 43 ) RICH HTML PDF (8995 KB) ( 38 )

	1283	Rapid Classification and Identification of Heavy Metal-Containing Electroplating Sludge by Combining EDXRF With Machine Learning
		LI Wei-yan¹, TENG Jing^2, ZHENG Zhi-hui^{3, 4}, SHI Jing-jing⁴, SHI Yao^4, LI Zhi-hong⁴, ZHANG Chen-mu⁴
		DOI: 10.3964/j.issn.1000-0593(2025)05-1283-07
		The rapid identification, classification, and pollution source tracing of hazardous wastes containing heavy metals is crucial to regional ecological and environmental quality supervision. This study used the energy-based X-ray fluorescence spectroscopy device (EDXRF) self-developed by the research group to collect spectral information of 8 different types of electroplating sludge from over 100 companies in Dongguan City. After spectral information noise reduction and data standardization, key classification factors were identified and used as input variables. The best method system for rapid X-fluorescence classification and identification of electroplating sludge containing heavy metals was determined through training and comparison of different machine models. The results show that the characteristic spectral line signals corresponding to the six metal elements of iron, copper, nickel, zinc, lead, and calcium can be used as a key factor to distinguish different types of electroplating sludge. Although random forest (RF), support vector machine (SVM), and linear discriminant (LDA) could achieve accurate classification and identification of electroplating sludge using X-ray fluorescence spectrum, only the RF model achieves 100% accuracy, precision, and sensitivity. The combination of machine learning and EDXRF technology can solve key problems such as the long, time-consuming, and poor timeliness of traditional chemical analysis methods for identifying hazardous wastes containing heavy metals. In the future, it will have broad application prospects in ecological environment monitoring and management such as rapid traceability of heavy metal pollution in soil and rapid identification of hazardous wastes containing heavy metals.
		2025 Vol. 45 (05): 1283-1289 [Abstract] ( 45 ) RICH HTML PDF (7022 KB) ( 102 )

	1290	Discriminative Study on Huzhu Qingke Liquor by Back Propagation Neural Network Combined With Ultraviolet-Near Infrared Fusion Spectroscopy
		ZHAO Yu-xia¹, ZHANG Ming-jin^{1, 3*}, WANG Ru¹, ZHANG Shi-zhi², YIN Bo^{1, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1290-10
		Chinese Huzhu Qingke Liquor is a protected geographical indication product, and it is of great significance for its accurate evaluation and classification. Due to the advantages of ultraviolet (UV) and near-infrared (NIR) spectroscopy, such as fast, accurate, non-destructive detection and no sample pretreatment, are widely used in food and other fields. In this study, a fast, nondestructive, and efficient discriminative classification model for Huzhu Qingke Liquor was established based on UV, NIR, and UV-NIR intermediate data fusion spectroscopy (UV-NIR) combined with theback-propagation neural network (BPNN) method. Since the unoptimized spectra are affected by noise and baseline drift due to the superimposed interference of spectral eigenpeaks, the spectra are denoised using four preprocessing methods, namely, standard normal variable transform (SNV), Savitzky-Golay smoothing (SG), first-order derivative (1D) and second-order derivative (2D). Further, relative to a single spectrum, the fused spectrum can complement the diversified spectroscopic information and improve the performance of the classification model, so the feature variables are selected by five variable screening methods, namely, competitive adaptive reweighted sampling (CARS), successive projection algorithm (SPA), principal component analysis (PCA), variable projection importance analysis (VIP), and variable combinatorial clustering analysis (VCPA) to achieve the optimization of model performance and the purpose of fusing the effective information of two spectra. Finally, the best method for establishing the BPNN model for single and fused spectra was selected. The results show that the classification model established by selecting 30 feature variables by SPA after SNV preprocessing for UV spectra has the best recognition effect, with a classification accuracy of 100%. The MSE value, R²_P, R(Train), R(Validation), R(Test) and R(All) were 0.018 0, 1, 0.928 3, 0.958 7, 0.913 0, and 0.929 7, respectively; PCA selected the NIR and UV-NIR after SG preprocessing with 84 and 106 The classification model built by feature variables had the best recognition effect, and the NIR spectral classification accuracy was 100%, with MSE value, R²_P, R(Train), R(Validation), R(Test)and R(All)of 0, 1.000, 1.000, 1.000，1.000 and 1.000. respectively, UV-NIR spectral classification accuracy was 100%, MSE, R²_P, R(Train), R(Validation), R(Test), and R(All) were 0.005 7, 1.000, 1.000, 0.987 1, 0.991 3 and 0.996 4, respectively; the fusion spectra can significantly improve the predictive ability and robustness of the classification model compared with the single-spectrum modeling, thus realizing the rapid and non-destructive analysis of Huzhu Qingke Liquor.
		2025 Vol. 45 (05): 1290-1299 [Abstract] ( 38 ) RICH HTML PDF (13898 KB) ( 16 )

	1300	Cotton Verticillium Wilt Severity Detection Based on Hyperspectral Imaging and SSFNet
		WU Nian-yi¹, CANG Hao¹, GAO Xiu-wen¹, LI Yong-quan¹, TAN Fei¹, DI Ruo-yu¹, RUAN Shi-wei¹, GAO Pan^1, LÜ Xin^2
		DOI: 10.3964/j.issn.1000-0593(2025)05-1300-10
		Verticillium wilt poses a severe threat to cotton yield and quality. Rapid and accurate detection of Verticillium wilt is essential for controlling cotton Verticillium wilt (CVW). Existing CVW detection methods mainly focus on the image or spectral level, overlooking the importance of feature fusion, which limits model performance. We propose a CVW grade detection method, spatial-spectral Fusion Network (SSFNet), to address this. First, we enhance the LAB color space, which is sensitive to pixel changes in infected plants, to enrich the feature representation of RGB images and use an improved ResNet network to build an image feature extraction module. Next, we construct a spectral feature extraction module based on the improved ResNet network and compare the performance of two common feature extraction methods: Least Absolute Shrinkage and Selection Operator (LASSO) and Principal Component Analysis (PCA). Finally, we build the feature fusion model SSFNet based on image and spectral level exploration. Experimental results show that SSFNet performs best compared to single data type features, with an F1 score of 95.96%, demonstrating the potential of image-spectral feature fusion methods combined with deep learning for CVW grade detection.
		2025 Vol. 45 (05): 1300-1309 [Abstract] ( 48 ) RICH HTML PDF (22496 KB) ( 26 )

	1310	Strawberry Defect Detection and Visualization Via Hyperspectral Imaging
		ZHAO Lu-lu^{1, 2}, ZHOU Song-bin^{1, 2}, LIU Yi-sen^{1, 2*}, PANG Kun-kun^{1, 2}, YIN Ze-xuan^{1, 2}, CHEN Hong^{1, 2}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1310-09
		Strawberries can be easily damaged during harvesting, transportation, storage, packaging, and sales. The damages and defects encountered include bruising, frost damage, and fungal infections, which can cause great economic losses to fruit farmers and sellers. Hyperspectral technology combines spectral sensing and machine vision to non-destructively detect various quality defects in fruits. However, there are currently two problems in modeling hyperspectral fruit detection: First, the input information is mainly based on average spectra, and the hyperspectral image information is not adequately utilized. Secondly, convolutional networks (CNN) have become the main focus of development in hyperspectral data processing. Still, CNNs have a relatively narrow domain of perception, and it is difficult to obtain long-term correlations for spectral segments or image information. To solve the above problems and accurately detect and recognize various strawberry defects, a spatial-spectral transformation network (SSTN) was proposed to classify the near-infrared hyperspectral data (900~1 700 nm) of four categories of strawberries (healthy, bruised, frost damaged, and infected). The SSTN uses the Vision Transformer (ViT) network as the main body and hyperspectral data patches with encoded position information are used as inputs to achieve “spectra-spatial” modeling. The model's internal multi-head attention mechanism can also capture long-distance spectral/spatial correlations. In the experiment, 502 strawberries were sampled, including 128 healthy, 128 bruised, 128 frost damaged, and 118 infected strawberries. The training and test sets were randomly divided according to a1∶1 ratio. Half of the data was used to train the model to classify defects, and the other half was used to test the model's performance. The results show that SSTN achieved a maximum classification accuracy of 99.20%. Compared with one-dimensional convolutional neural network (1D-CNN), two-dimensional convolutional neural network (2D-CNN), and convolutional network with attention mechanism (CBAM-CNN), our SSTN model achieved accuracy improvements of 3.8%, 3.3%, and 1.5%, respectively. The trained 2D-CNN, CBAM-CNN, and SSTN models were combined with Score-CAM for visualization to further visualize the location of various strawberry defects. The results of defect visualization show that the convolutional attention mechanism in CBAM-CNN can improve the accuracy of the defect location, while the SSTN model with multi-head attention mechanism combined with Score-CAM had the best visualization performance, which can be used to accurately display the location and outline of the defect shape. This study provides a reference for establishing a fast, non-destructive, and automatic detection method for strawberry defects.
		2025 Vol. 45 (05): 1310-1318 [Abstract] ( 49 ) RICH HTML PDF (19209 KB) ( 32 )

	1319	Study on the Coupling of Spectral Method With Sensitizer to Reduce the Detection Limit of Trace Indium in Copper Smelting Dust
		QU Wei^{1, 2}, LI Zi-shang^{1, 2}, LI Qian^{1, 2}, ZHANG Hong-zhi^{1, 2}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1319-06
		For the determination of trace indium in copper smelting dust, the standard method adopts flame atomic absorption spectroscopy, and the determination range of indium content is 0.020 0%～0.100%, which cannot achieve the determination of indium content below 0.020 0%. To determine indium content below 0.020 0%, reference is made to the national standard “Chemical Analysis Method for Zinc Concentrates”. However, during the testing process, trace amounts of indium need to be extracted with butyl acetate and back extracted with hydrochloric acid solution. The steps are cumbersome, the operation is complex, and the analysis is time-consuming, which cannot be quickly combined with scientific research in mineral processing and metallurgy. This article studies the coupling of spectroscopy with sensitizer acetone (Ac) in the presence of surfactant Triton X-100 (OP) in a hydrochloric acid medium, which suppresses the interference of lead, zinc, copper, bismuth, and other elements and reduces the detection limit of trace indium in copper smelting dust. Mechanism studies have found that under certain conditions, indium ions, acetone, and Triton X-100 form ternary complexes, which synergistically enhance the atomic absorption spectra of indium, improve the intensity of indium atomic absorption spectra, and increase their sensitivity-using multiple linear regression analysis to fit the concentration relationship between absorbance (A) and indium (In), acetone (Ac). The P values of c_In, c_Ac and c_Op in the fitting results are all less than 0.05, indicating that the three significantly impact absorbance. Using multiple linear regression analysis to find the extremum method, the molar ratio of the ternary complex was calculated to be 1∶1∶2. In addition, this article studied the spectral characteristics of indium determined by flame atomic absorption spectroscopy, the additional amount of acetone and Triton X-100, the selection and addition amount of acid types, acetylene flow rate, and the influence of coexisting ions. Comparative experiments were conducted using the extraction method, and mathematical statistics analyzed the results. The conclusion was drawn that this method has good accuracy and precision and can be used for the analysis and detection of trace indium in copper smelting dust; The detection limit of the method was determined using reagent blank, and at a confidence level of 99%, the detection limit of the method was 0.001 0%, which is 20 times lower than the industry standard method.
		2025 Vol. 45 (05): 1319-1324 [Abstract] ( 51 ) RICH HTML PDF (2873 KB) ( 10 )

	1325	Synthesis and Characterization of Titanium Dioxide Coated Carbon Dots and Their Applications in Fingerprint Development
		MA Rong-wei, WANG Meng^*, LI Jie, XU Zhi-ze, LI Ming, YUAN Chuan-jun
		DOI: 10.3964/j.issn.1000-0593(2025)05-1325-09
		Latent fingerprints (LFs) are commonly encountered at crime scenes. Developing the LFs clearly is the precondition for further analysis and identification. In this work, titanium dioxide-coated carbon dots (CDs@TiO₂) fluorescent nanosuspensions were prepared and used for high-quality LF development. Firstly, carbon dots (CDs) were synthesized via a solvothermal approach using citric acid and urea as raw materials and N,N-dimethylformamide as solvent. Then, CDs@TiO₂ nanomaterials (NMs) were formed by coating CDs with a layer of TiO₂ shell based on the ammonia-catalyzed hydrolysis of titanium butoxide, and the synthesis conditions were optimized. The optimized synthesis conditions were as follows: the amount of CDs, tetrabutyl orthotitanate, water, and ammonium hydroxide was 3.0, 1.5, 1.5, and 0.1 mL, respectively, the reaction temperature was 50 ℃, and the dropping period was 30 min. After that, the morphology, composition, structure, and optics properties of CDs and CDs@TiO₂ NMs were characterized. Characterization results showed that, CDs were near spherical with an average diameter of 7.54 nm, they could give characteristic Raman scattering peaks as well as infrared absorption peaks of CDs, and possessed the crystal structure of hexagonal graphite, their UV absorption peak was at 343 nm, and their maximum fluorescence excitation and emission wavelength was at 450 and 567 nm respectively; CDs@TiO₂ were irregularly spherical with an average diameter of 114.85 nm, they could give characteristic Raman scattering peaks as well as infrared absorption peaks of both CDs and TiO₂, and possessed the crystal structure of both hexagonal graphite and tetragonal rutile TiO₂, their UV absorption peak was at 321 nm, and their maximum fluorescence excitation and emission wavelength was at 387 and 529 nm respectively. Finally, CDs@TiO₂ NMs were made into nanosuspension for developing LFs via hydrophobic interaction. The development conditions were optimized, and the results of LF development were investigated in detail. The optimized development conditions were as follows: the concentration of sodium dodecyl sulfate and choline chloride was 1.0‰~2.0‰ and 4.0‰~6.0‰ respectively, and the developing time period was 10~20 s. Experimental results showed that, the LFs could emit bright blue fluorescence under 490 nm excitation, which exhibited coherent and clear papillate ridges and distinct minutiae. Our proposed method based on CDs@TiO₂ fluorescent nanosuspensions could develop the LFs on common smooth and non-porous substrates with high quality, possessing enough contrast, high sensitivity, good selectivity, and wide applicability.
		2025 Vol. 45 (05): 1325-1333 [Abstract] ( 52 ) RICH HTML PDF (42170 KB) ( 10 )

	1334	Study on the Influence of Slurry Solid Concentration and Particle Size on LIBS Measurement Signals and Characterization Methods
		YU Tong¹, YU Hong-xia¹, ZHANG Peng^{2, 3}, SUN Lan-xiang^{2, 3}, CHEN Tong^{2, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1334-07
		Laser-Induced Breakdown Spectroscopy (LIBS) technology can directly analyze online slurry. However, due to the complex matrix composition of slurry, variations in solid particle concentration and particle size can affect the spectral signal, making the relationship between the spectral signal and mineral chemical composition difficult to characterize. Therefore, studying how slurry solid phase concentration and particle size influence the spectral signal is crucial. In this study, SiO₂ powder of various particle sizes was mixed with different amounts of water to create simulated slurry samples with varying solid-phase concentrations and particle sizes. LIBS spectral signal acquisition and data analysis were conducted to systematically investigate the impact of solid-phase concentration and particle size on the intensity of characteristic spectral lines of elements in both the solid and liquid phases. Firstly, the Pearson coefficient was used to quantitatively evaluate the impact of solid-phase concentration and particle size on the LIBS spectral signal. The results showed that, under the same particle size, the correlation coefficient between solid-phase concentration and the full spectrum value and the intensity of characteristic spectral lines of various elements was mostly above 0.9, indicating that overall spectral intensity increases with the solid-phase concentration. Under the same solid-phase concentration, the correlation coefficient between the principal components of particle size and the intensity of silicon element characteristic spectral lines was around 0.99, indicating that the characteristic spectral lines of elements in solid particles weaken as the particle size increases. Furthermore, a model was established to describe the relationship between spectral intensity solid-phase concentration and particle size when both parameters change simultaneously. The model's goodness of fit (R²) was used to analyze the impact of solid-phase concentration and particle size on the LIBS spectrum. The results indicated that only the solid-phase particle size representation model for silicon element characteristic spectral line intensity achieved goodness of fit above 0.9, suggesting that neither solid-phase concentration nor particle size alone can fully and accurately reflect changes in the intensity of characteristic spectral lines of elements in the samples. The model that simultaneously characterizes spectral intensity based on both solid-phase concentration and particle size achieved a goodness of fit exceeding 0.9 for the characteristic spectral lines of all elements, indicating that the influence of solid-phase concentration and particle size on the intensity of characteristic spectral lines is coupled and requires comprehensive characterization through multi-source information integration. These findings provide a systematic analytical foundation for further research on enhancing the stability and accuracy of quantitative LIBS analysis based on multi-source information fusion.
		2025 Vol. 45 (05): 1334-1340 [Abstract] ( 37 ) RICH HTML PDF (9397 KB) ( 15 )

	1341	Study on Quantitative Analysis Method of TDLAS Intravenous Drug Concentration Based on ECA-1D-CNN
		ZHU Yong-bing¹, CAI Yu-qin¹, JIANG Li-yao¹, LEI Chun¹, TENG Long¹, WANG De-wang³, TAO Zhi^2*
		DOI: 10.3964/j.issn.1000-0593(2025)05-1341-07
		The quantitative analysis of intravenous drug solute concentration has always been the research hotspot of drug detection in static dispensing centers. Still, the conventional means of mixing and reviewing are operated manually. There are problems such as limited control of the concentration of the drug solution, laborious pressure of manual review, and inefficiency, so it is crucial to propose an accurate, convenient, and non-destructive detection method for intravenous drug solute concentration. Due to the limitations of traditional near-infrared spectroscopy for the detection of low-concentration liquids, based on tunable laser absorption spectroscopy (TDLAS) technology, a quantitative detection model of glucose mixed solution concentration based on efficient attention mechanism one-dimensional convolutional neural network (ECA-1D-CNN) was investigated. In order to detect the low concentration of glucose mixed solution, based on the TDLAS technology, the 980 nm band with the highest light intensity absorption rate was selected as the laser light source, and through the photoelectric sensor, the transmitted light intensity signal of the drug concentration was acquired, which was demodulated into the second harmonic signal by the phase-locked amplification module to obtain a total of 600 self-constructed datasets of different concentrations, and the samples were divided into training and testing sets in the ratio of 8∶2. Aiming at the second harmonic signal of the transmitted light intensity of 600 drug concentrations as the research object, a glucose mixed solution concentration detection model based on the one-dimensional convolutional neural network model with efficient attention mechanism (ECA-1D-CNN) is proposed, with a total of four convolutional layers, all of which are activated by the Relu activation function, and a BN layer is added after each convolutional layer, a pooling layer is added after every two convolutional layers, and a pooling layer is added after the 2nd pooling layer, and a pooling layer is added after the 2nd pooling layer, and a pooling layer is added after the 2nd pooling layer. Adding 1 ECA after the 2nd pooling layer can help the network model to learn the relationship between features better, reduce the number of parameters, and improve the robustness of the model. First, to highlight the advantages of the 1D-CNN model, the same original dataset is used to model PCR, SVR, and PLSR and compare the prediction effects of the 4 different models. Second, based on six different data preprocessing, the ECA-1D-CNN model was compared with the 1D-CNN model to analyze the generalization ability of the prediction model by using the coefficient of determination R², the absolute error MAE, and the root-mean-square error RMSE as the evaluation indexes. The results showed that the ECA-1D-CNN model under SG+Normalization preprocessing was the most effective, which was able to effectively predict the concentration of glucose mixed solution from 6 to 30 mg·100 mL^-1, and the R² of the model's training set could reach 0.998, the MAE 0.295, and the RMSE 0.343, and the R² of the test set could reach 0.993, MAE of 0.498, RMSE of 0.691. The proposed method can accurately predict the concentration of intravenous drug solutes, which provides a new idea and an application value for the nondestructive testing of intelligent static dispensing centers.
		2025 Vol. 45 (05): 1341-1347 [Abstract] ( 50 ) RICH HTML PDF (7660 KB) ( 11 )

	1348	Rapid Quantification of Illegal Addition of Lambda-Cyhalothrin in Bacillus Thuringiensis Preparations by Infrared Spectroscopy
		CHEN Yue-fei, XIA Jing-jing, WEI Yun, XU Wei-xin, MAO Xin-ran, MIN Shun-geng^, XIONG Yan-mei^
		DOI: 10.3964/j.issn.1000-0593(2025)05-1348-07
		Rapid quantitative determination of the illegal addition of lambda-cyhalothrin in Bacillus thuringiensis preparations was carried out using attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR) coupled with partial least squares (PLS) method. Three different sources of Bacillus thuringiensis preparations with different masses of 95% (w/w) lambda-cyhalothrin prodrug were added to prepare 153 mixed samples, concentrations ranging from 0.00% to 5.00%, acetonitrile was used as the extractant, and less extractant was used to increase the concentration of lambda-cyhalothrin in the extract to enhance the infrared absorption. Three pretreatment methods (smoothing, standard normal variation (SNV) and multiple scattering correction (MSC)) and six variable selection algorithms(uninformative variable elimination (UVE), interval partial least squares (iPLS), moving window partial least squares (MWPLS), competitive adaptive reweighted sampling (CARS), the bootstrapping soft shrinkage (BOSS) and interval combination optimization (ICO)) , were used to examine the effects of different pretreatment methods and variable selection methods on the model results. Among them, the MSC pretreatment method combined with the BOSS algorithm obtained the optimal model results, with this method, RMSECV=1.18×10^-3, R²_cv=9.94×10^-1, RMSEP=1.01×10^-3, and R²_pre=9.93×10^-1. For samples with lambda-cyhalothrin concentrations ranging from 0.10% to 5.00%, the average relative error of external test samples was 4.44%, and for samples with concentrations ranging from 2.00% to 5.00%, the average relative error of external test samples was only 2.64%. This method could be applied to rapidly detect the illegal addition of lambda-cyhalothrin in Bacillus thuringiensis preparations.
		2025 Vol. 45 (05): 1348-1354 [Abstract] ( 38 ) RICH HTML PDF (4728 KB) ( 22 )

	1355	Examination of the Correlation Between Shortwave Infrared Spectra and Mineral Geochemical Characteristics of Muscovite in the Pegmatite-Type Lithium Deposit in Lijiagou, Western Sichuan
		HAN Jing-rui¹, RAN Feng-qin^1*, PENG Bo¹, CHEN Ran³, TANG Bo², FENG Da-bo¹, YANG Yang¹, ZHAO Yuan¹, GU Chun-jin², CHEN Hao⁴, ZHAO Heng-bing⁵
		DOI: 10.3964/j.issn.1000-0593(2025)05-1355-09
		The Liajiagou deposit, a massive granite pegmatite-type lithium deposit, is located in the western part of Sichuan Province. Currently, the shortwave infrared spectroscopic properties of muscovite in the Liajiagou deposit remain unknown, and a deeper understanding of the correlation between its chemical composition and spectral characteristics is required. This study examines multiple generations and types of muscovite in the pegmatite of the Li Jiagou ore deposit. Microstructural observations, shortwave infrared (SWIR) spectroscopy, and electron probe microanalysis (EPMA) were performed. The muscovite in the pegmatite was classified into three generations: ①Primary muscovite (MS), formed during the magmatic stage, is characterized by high concentrations of aluminum (Al) and potassium (K). The absorption peak of muscovite at 2 200 nm (Pos2200) displays a limited variation range, ranging from 2 201 to 2 202.8 nm. The absorption depth at this peak (2200 Dep) often remains below 0.25. Additionally, the illite crystallinity index (IC value) typically falls within a narrow range of 1 to 1.5. ② During the magma-hydrothermal transition stage, transition muscovite (TM) becomes enriched in Li elements as a result of the exsolution of volatile-rich fluids. The Li₂O content in TM varies between 2.53% and 6.22%. Additionally, Pos2200 demonstrates a concentration peak centered at 2 200～2 201 nm. Most 2200 Dep values are below 0.5, whereas IC values remain consistently high, often exceeding 4. ③Hydrothermal muscovite (HM), predominantly formed via fluid metasomatism, exhibits the shortest Pos2200 values, with 2200 Dep readings mostly below 0.4 and IC values ranging from 2 to 4. The Al—OH absorption peak of muscovite in the Lijiagou pegmatite indicates variations in the elemental composition, including Si, Al, Fe, Li, and other constituents. During the transition from the muscovite (MS) stage to the topaz-muscovite (TM) stage, changes in elemental composition and substitution mechanisms primarily influence the shift in muscovite's characteristic absorption peak wavelength. In contrast, temperature becomes the primary determinant during the hydrothermal (HM) stage. Comprehensive research highlights the potential of the characteristic absorption wavelength (2 200～2 201 nm) of the muscovite Al—OH peak, coupled with an illite crystallinity value greater than 4, as effective indicators for mineral exploration in the Lijiagou pegmatite.
		2025 Vol. 45 (05): 1355-1363 [Abstract] ( 45 ) RICH HTML PDF (28806 KB) ( 24 )

	1364	Background Subtraction Method for Soil XRF Spectrum Based on PIEspline
		LI Tang-hu¹, GAN Ting-ting^{2, 4}, ZHAO Nan-jing^{1, 2, 3, 4, 5}, YIN Gao-fang^{2, 4, 5}, YE Zi-qi^{2, 3, 4} , WANG Ying^{2, 3, 4}, SHENG Ruo-yu^{2, 3, 4}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1364-09
		X-ray fluorescence (XRF) spectroscopy is a crucial technique for the rapid on-site detection of heavy metals. However, when applied to soil heavy metal analysis, the presence of high-intensity and complex background spectra due to soil matrix effects significantly hinders the accurate acquisition of characteristic spectral peaks and the precision of quantitative analyses. To address this issue, this paper proposes a background subtraction method for soil XRF spectra, combining peak-valley recognition using an extremum method with penalized correction for cubic smoothing spline fitting, termed PIEspline. Initially, the extremum method identifies peak and valley points in the complete soil XRF spectra to extract data points representative of the background. These points are then used to fit a cubic smoothing spline curve with penalized corrections, forming the background baseline and thus enabling the subtraction of complex backgrounds from soil XRF spectra. The performance of the PIEspline method is further validated by comparing it with three traditional spectral background subtraction methods: adaptive iteratively reweighted penalized least squares (airPLS), iterative wavelet transform (IWT), and statistical sensitive nonlinear iterative peak-clipping (SNIP). The results indicate that, for simulated soil XRF spectra, the root mean square errors (RMSE) between the background spectra obtained by the PIEspline method and the true background spectra are 0.425 8 and 0.644 1, respectively, which are lower than those of the other three methods. Additionally, the PIEspline demonstrates the fastest background subtraction efficiency. For three different soil types cinnamon soil, saline-alkali soil, and loess and three different soil uses agricultural, industrial, and construction the average relative error of fluorescence intensity at 10 characteristic valley points in the XRF spectra fitted by PIEspline is 10.87%. Compared to the traditional methods, this error is reduced by 84.88%, 76.30%, and 16.51%, respectively. Furthermore, for quantitative analysis of Cr, Pb, and Cd in the aforementioned six soil types, the average relative errors using PIEspline are 4.01%, 2.50%, and 5.20%, respectively, representing reductions of 22.39% to 84.07%, 60.15% to 71.92%, and 79.18% to 84.07% compared to airPLS, IWT, and SNIP. Notably, the PIEspline method exhibits minimal fluctuation in relative error when soil type and use change, showcasing superior stability. This suggests that the PIEspline method offers the best applicability for simultaneous XRF quantitatively analyzing multiple heavy metals across various soil types and uses. Therefore, the PIEspline method proposed in this study enables precise background subtraction of XRF spectra from different soil types and uses, improving heavy metal XRF quantitative analysis accuracy. This research provides a vital methodological foundation for the rapid and accurate on-site detection of soil heavy metals using XRF.
		2025 Vol. 45 (05): 1364-1372 [Abstract] ( 47 ) RICH HTML PDF (6256 KB) ( 10 )

	1373	Analysis and Research on Polychrome Pigments for the Statues of Yuanjue Cave, Anyue Grottoes
		ZHOU Wei-qiang¹, LIU Ning¹, HE Jing², CHEN Hui-li³, LEI Yu³, RUAN Fang-hong³, HOU Jing-min⁴
		DOI: 10.3964/j.issn.1000-0593(2025)05-1373-10
		The statue of Yuanjue Cave in Anyue Grottoes was built in the Tang Dynasty, flourished in the Five Dynasties and Song Dynasty, which has the characteristics of inheriting Longmen and Yungang Grottoes and inspiring Dazu Rock Carving. It occupies an important position in the history of Chinese Buddhist grotto art. Due to the humid semi-open-air natural environment, improper restoration, and other human factors, a large area of pigment chalking, peeling, and flaking appeared on the statue's surface, and some were covered with modern pigments. However, there is a lack of relevant scientific analysis. To realize the conservation and restoration of the painted statues, it is necessary to analyze the composition and process of the original painted pigments. In the research process, we mainly use the combination of microscopic analysis, scanning electron microscope-energy spectrum analysis, and micro-area Raman spectroscopy to comprehensively determine the pigment composition and hierarchical structure and summarize the painting process of AnyueYuanjue Cave. At the same time, we will discover the phenomenon of pigment repainting and initially explore the relationship between pigment repainting and pigment flaking disease. The experimental results show that the color painting of Yuanjue Cave is based on a white power layer, and gypsum or calcite is used as the base material. This is consistent with the period of cave excavation and restoration. The surface decoration materials use a variety of traditional pigments: red pigments are hematite and vermilion, orange pigments are minium, green pigments are malachite, blue pigments are ultramarine, black pigments are carbon, and gold is gold. In addition, combined with the microscopic section and Raman results, a variety of modern synthetic pigments are also found in K7, K9, and K42, such as lead-chrome yellow, Prussian blue, and lavendulan (the decomposition of emerald green), etc., indicating that the statue was later painted. Multiple repainting increases the overall thickness of the pigment layer, and differences in the properties of the pigment binding medium may exacerbate the production of pigment flaking disease. This study provides a scientific basis for identifying colored pigment and the protection and restoration of Yuanjue Cave.
		2025 Vol. 45 (05): 1373-1382 [Abstract] ( 53 ) RICH HTML PDF (70497 KB) ( 23 )

	1383	Study on the Fluorescence Spectral Characteristics of Egyptian Blue Mineral Pigment and Their Application in Development of Latent Fingerprints
		DAI Xue-jing¹, TANG Cheng-qing², LI Yun-peng¹, SONG Jia¹
		DOI: 10.3964/j.issn.1000-0593(2025)05-1383-06
		The development and identification of the fingerprints left by the suspects are the important basis for detecting cases. The majority of commercially available luminescent fingerprint powders can usually fluoresce upon excitation with ultraviolet (UV) light or blue and green light. Still, multi-colored and patterned backgrounds have intrinsic formulations (inks, binders, coatings, etc.), which also cause them to fluoresce within the same spectrum area. This creates a lack of distinguishing ability between the fluorescence of the developed latent fingerprint and the background. In addition, -most fluorescent powders have seriously damaged the long-term health of forensic technicians. Therefore, it is urgent to find a kind of non-toxic fluorescent fingerprint power suitable for use on substrates that are typically considered extremely difficult to treat. In this paper, the fluorescence properties of the natural Egyptian blue mineral pigment, which was harmless and cheap, were studied. The micro-morphology, crystal structure, luminescence properties, and surface functional groups of the Egyptian blue mineral pigment were characterized by scanning electron microscopy, X-ray diffractometer, fluorescence spectrophotometer, and Fourier transform infrared spectrometer. The Egyptian blue mineral pigment had good dispersion with anaverage diameter of about 1 μm, and its main component was a mixture of CuO and Cu₂O. Under the excitation of 780 nm infrared, it could give strong emissions at the wavelength of 823 nm infrared. It was a silicate mixture containing copper ions. Finally, the Egyptian blue mineral pigment was used to develop the latent fingerprints on extremely troublesome substrates commonly encountered at crime scenes. The experimental results showed a high contrast between the fingerprint and the background with fine ridge detail and less interference. In addition, the excitation and emission spectra of the Egyptian blue mineral pigment were in the near-infrared region, which filled the blank of the method of developing fingerprints with fluorescence powders.
		2025 Vol. 45 (05): 1383-1388 [Abstract] ( 45 ) RICH HTML PDF (34306 KB) ( 13 )

	1389	Analysis of Pigments of Ming Dynasty Polychrome Paintings Composition in the Juehuang Hall of the Mingjiao Temple
		ZHANG Wen-jie¹, ZHANG Yu², CAO Zhen-wei³, HAN Xiang-na^1*, GUO Hong¹
		DOI: 10.3964/j.issn.1000-0593(2025)05-1389-06
		Mingjiao Temple is located in Chengdu, Sichuan Province. The Juehuang Hall is the only surviving structure of the original Mingjiao Temple complex. The main part of the hall was constructed during the early Hongwu period of the Ming Dynasty (1368—1382) and was completed no later than the first year of the Chenghua reign (1465). The interior of Juehuang Hall retains a significant number of Ming Dynasty murals, which exhibit the distinctive characteristics of the official architectural style of the Northern country. These murals are a rare example of Ming Dynasty architectural painting and hold significant research value. Previous research on the paintings in the Juehuang Hall has focused on their form and aesthetic style, with no scientific analysis of their production techniques and materials. This study employs a high-resolution digital microscope, laser Raman spectrometer, scanning electron microscope, and energy dispersive spectrometer to analyze and identify the pigments from the paintings in the Juehuang Hall. The results indicate that the green pigments in the Mingjiao Temple's Juehuang Hall paintings are malachite and copper chloride, the red pigments are iron oxideand red lead, the white pigment is lead white, the blue pigment is indigo, and the black pigment is carbon black. Furthermore, the eaves paintings within the Juehuang Hall exhibit multi-layered paintings and the practice of mixing pigments for color adjustment. Synthetic ultramarine and Paris green are absent in these eaves paintings, which were commonly used in the Qing Dynasty's middle and later periods. The painting technique, characterized by the direct application of pigments onto the wooden components without a preparatory ground layer, suggests that these paintings are likely remnants from the Ming Dynasty, aligning with the documented period of the paintings' creation. This study is the first scientific analysis of the production techniques and materials of the Juehuang Hall paintings, and the preliminary findings have enriched the understanding of the application of pigments in Ming Dynasty architectural paintings, providing a reference for subsequent research and conservation efforts.
		2025 Vol. 45 (05): 1389-1394 [Abstract] ( 41 ) RICH HTML PDF (48590 KB) ( 13 )

	1395	Distribution and Content of Iron Sulfides in the Wood of Nanhai Ⅰ Shipwreck
		WANG Xue-yu, LI Nai-sheng^*, DU Jing
		DOI: 10.3964/j.issn.1000-0593(2025)05-1395-08
		The Nanhai Ⅰ Shipwreck is China's first ancient shipwreck to be salvaged from the ocean. It is renowned for its well-preserved condition, large size, and abundant cultural relics. The wood of the Nanhai Ⅰ Shipwreck contains a considerable amount of iron sulfides, which threaten the shipwreck's long-term and safe preservation. However, a significant gap exists in studying the iron sulfides in the shipwreck's wood. This study selected three typical cabins of the Nanhai Ⅰ shipwreck (No.4, No.7, and No.11), which had significant differences in the quantities of iron cargo loaded, as the research subjects. Wet chemistry methods, optical microscopy (OM), scanning electron microscopy (SEM), X-ray diffraction (XRD), and inductively coupled plasma emission spectrometer(ICP) were used to analyze the wood degradation of different cabins, as well as the distribution, morphology, and composition of the iron sulfides, and sulfur and iron content. The results showed that the degree of wood degradation varied among different cabins. The ratio of holocellulose content to lignin content (H/L) of the wood in cabins No.4 and No.7 was relatively low, indicating a higher degree of degradation. In contrast, the H/L value of the wood in cabin No.11 was the highest, indicating the lowest degree of degradation. The inorganic sediments in the wood of the Nanhai Ⅰ shipwreck are primarily iron sulfides, mainly including pyrite (FeS₂), hydroxylated iron oxide (FeOOH), and siderite (FeCO₃). These compounds are distributed within the wood cell structures such as tracheids and wood rays, and are attached to the inner layer of cell walls. Due to the ironware's loading conditions and the wood's preservation state, the iron content in the wood of the Nanhai Ⅰ shipwreck is relatively high and varies significantly among different cabins. The iron content in the wood of cabin No. 11 is mainly below 2.5%, while in cabins No.4 and No.7, it is mainly between 1%~5% and between 0.3%~30%, respectively. The sulfur element primarily originates from the degradation reaction of marine microorganisms and diffuses into the wood as hydrogen sulfide gas. Then it reacts with lignin to produce mercaptan and accumulates in the wood. Consequently, the sulfur content is relatively low, and the differences between cabins are minor, mainly concentrated between 5% and 10%. Based on this study, the degradation of wood and the distribution and content of iron sulfides in the wood of different cabins of the Nanhai Ⅰ shipwreck were investigated. The findings can guide the removal of iron sulfides from the wood of the Nanhai Ⅰ shipwreck.
		2025 Vol. 45 (05): 1395-1402 [Abstract] ( 38 ) RICH HTML PDF (29928 KB) ( 12 )

	1403	Scientific Research on the Chemical Composition and Making Process of a Batch of Ancient Glass Jue
		HUANG Jue-wei^{1, 2}, DONG Jun-qing^{1, 2}, LIU Song¹, YUAN Yi-meng¹, LI Qing-hui^{1, 2*}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1403-13
		Wearing ornaments is a traditional custom that has been popular from ancient to present. As a type of ornament, Jue is one of the traditional types of jade artifacts in ancient China, which profoundly influenced Southeast Asia. Jue has diverse materials and shapes, has significant historical and regional characteristics, and is one of the important carriers for technological and cultural exchange. Up until now, research on ancient Asian ornaments has mainly involved gemstones and glass beads, while there are few reports on glass Jue. Studying the chemical compositions and trace element characteristics of glass Jue is expected to provide scientific references for the production era of glass Jue found in Southeast Asia and cultural exchanges along the Maritime Silk Road. In this paper, 44 glass Jue found in Southeast Asia have been scientifically analyzed using energy dispersive X-ray fluorescence spectroscopy (EDXRF), laser ablation inductively coupled plasma mass spectrometry(LA-ICP-MS), optical coherence tomography (OCT), optical microscopy (OM) and confocal laser Raman microspectroscopy. The Jue samples ' chemical compositions, phase components, internal physical structures, and surface micromorphology are obtained. According to the main flux content of these glass Jue samples, they can be classified into three glass types: potash glass, soda-alumina glass, and potash-lead glass. 42 samples belong to potash glass, and the 2 samples left belong to soda-alumina glass and potash-lead glass, respectively. Two subgroups, the m-K-Al subgroup and the m-K-Ca-Al subgroup, are found for potash glass. Only 4 potash glass samples belong to the m-K-Ca-Al subgroup; the others are the m-K-Al subgroup potash glass. Drawing, casting, and cold-working techniques were applied during glass making. Most of the blue-green, green, and yellow-green glass Jue samples are mainly colored with iron ions, and some of them are colored with both iron and copper ions. The blue and blue-black samples are colored with cobalt ions; the black sample SEAG-005 is colored with manganese ions. The potash-lead glass SEAG-031 is opacified by lead-tin yellow. Combining the glass types, trace element characteristics，shapes, and distributions of Jue, the origins of the raw materials, and production centers for the glass Jue samples are discussed. These glass Jue samples in Southeast Asia witnessed the spread of traditional Chinese culture to Southeast Asia through the Maritime Silk Road and its integration with local technology and culture between 500 BCE and 500 CE. The results of this paper provide a preliminary basis for the cooperation research between China and other countries along the Maritime Silk Road.
		2025 Vol. 45 (05): 1403-1415 [Abstract] ( 56 ) RICH HTML PDF (20124 KB) ( 17 )

	1416	Spectroscopic Analysis of Compact Binary Candidate LAMOST J051402.68+172659.7
		WANG Qi¹, YANG Hai-feng^2, CAI Jiang-hui^3
		DOI: 10.3964/j.issn.1000-0593(2025)05-1416-06
		Compact binary systems usually have faster orbital motion, more frequent light variation effects, and the rapid evolution of companion stars, so they are important for studying interstellar physical processes and stellar evolution. Photometric images and light curves are the most common approaches to identifying binary stars. Suppose the current observation technology cannot distinguish binary objects' brightness and position difference. In that case, the spectral type is also one of the important methods used to distinguish the composition of the companion star. This paper uses the spectral binary star analysis method based on a rough set and cluster voting mechanism to analyze the spectral characteristics and identify the outlier LAMOST J051402.68+172659.7. First, for the spectral binary stars (DoubleStar) released by LAMOST DR10, human inspecting was conducted on the 17 target spectra with the lowest number of votes in its multiple clustering results. In addition to the low signal-to-noise ratio of most spectra, the two observed spectra of LAMOST J051402.68+172659.7 show great difference; For the two observations of the target, its spectrum once presents F-type and once presents F+M type. After excluding the pollution of neighboring optical fibers and other targets in the target's environment, it is determined that the real components of the target are displayed in the two observation spectra, indicating that the two spectral components come from two stellar targets. The radial velocity difference between the two observations is about 20.3 km·s^-1, and the observation interval is 3 days, indicating that the rotation period is less than 6 days. Considering that the conditions for the simultaneous occurrence of multiple target components in the same spectrum are extremely strict, the spectral types are very different, and the luminance (flow) is on the same scale, the spectral components, imaging, and light variation are analyzed in depth in this paper. From the image sequence of ZTF photometry, it can be detected that the size of the target contour presents periodic changes. In contrast, no obvious periodic changes are found on the light curve of ASAS-SN and ZTF. At the same time, because the distance between the two celestial bodies is very close or the projection is covered, the characteristics of the point source (the boundary is slightly irregular) cannot be identified from the SDSS and 2MASS spectrophotometry images, so the orbit between the two stars of the binary star system may be relatively close and the radius is small. Balmer weak emission lines (Hα, Hβ, Hγ, Hδ) and forbidden lines ([NII]λ6550, [SII]λλ6718, 6733, [OII]λ3728) are present in its spectrum, and the target is not in the planetary nebula or HII region, which is speculated to be the tidal effect and material exchange caused by the interaction of companion stars. This impacts the star's surface and atmosphere, creating emission lines. In addition, the target spectrum also shows extremely weak emission line components of suspected background galaxies, which cannot be distinguished from imaging and are presumed to be data processing remnants.
		2025 Vol. 45 (05): 1416-1421 [Abstract] ( 33 ) RICH HTML PDF (20684 KB) ( 9 )

	1422	Spectral Prediction of Soil Fertility Attributes in Typical Croplands of Sanjiang Plain Based on Band Selection
		YAO Cheng-shuo^{1, 2}, WANG Chang-kun^{1, 2*}, LIU Jie^{1, 2}, GUO Zhi-ying^{1, 2}, MA Hai-yi^{1, 2}, YUAN Zi-ran^{1, 2}, WANG Xiao-pan^{1, 3}, PAN Xian-zhang^{1, 2}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1422-10
		The Sanjiang Plain is an important grain production area in the black soil region of Northeast China. However, since its reclamation, the soil fertility of cultivated lands has declined significantly. Traditional chemical measurement methods are inefficient and difficult to meet the needs of rapid and accurate monitoring of soil fertility attributes. Spectral technology has the potential to predict soil fertility. Still, few existing studies have targeted multiple soil fertility attributes simultaneously, and the prediction accuracy of some soil fertility attributes is relatively low. Therefore, this study took the typical cropland area of the Sanjiang Plain, Youyi Farm, as the study area. We utilized visible and near-infrared spectroscopy, combined with four spectral preprocessing methods, including SG (Savitzky-Golay) spectral smoothing, first-order derivation, standard normal variate transformation, and multiplicative scatter correction, as well as the competitive adaptive reweighted sampling (CARS) band selection algorithm. The partial least squares regression model was employed to simultaneously predict four key soil fertility attributes: soil organic matter (SOM), total nitrogen (TN), total phosphorus (TP), and total potassium (TK). The study aimed to explore the potential of spectral prediction for multiple soil fertility attributes and investigate the role of variable selection in improving prediction accuracy. The results showed that: (1) When using the full spectral range (400~2 400 nm) without variable selection, the prediction accuracy of SOM and TN was relatively high, with cross-validation R² values ranging from 0.85 to 0.89 and 0.86 to 0.89, respectively. The prediction accuracy of TK was also relatively high, with R² ranging from 0.63 to 0.72, but the prediction accuracy of TP was lower, with R² ranging from 0.08 to 0.34. (2) After CARS band selection, the prediction accuracy of all four soil fertility attributes improved, with the largest improvement found in TP. The optimal cross-validation R² was 0.97, 0.96, 0.82, and 0.92 for SOM, TN, TP, and TK, respectively. (3) The CARS variable selection method identified the spectral bands corresponding to the characteristic functional groups related to SOM and TN. The prediction of TN utilized both its relationship with SOM and its intrinsic characteristic bands. The prediction of TP mainly relied on soil spectral information, while the prediction of TK utilized both soil spectral information and its relationship with SOM and TN. This study demonstrated the potential of spectral technology for simultaneously predicting multiple key soil fertility attributes in the typical cropland area of the Sanjiang plain and found that variable selection can significantly improve the prediction accuracy of soil attributes(TP) that do not have obvious spectral characteristics, providing a methodological reference for rapid monitoring of soil fertility in black soil regions.
		2025 Vol. 45 (05): 1422-1431 [Abstract] ( 44 ) RICH HTML PDF (20608 KB) ( 14 )

	1432	Spectral Characteristics of Dissolved Organic Matter in Black Soil Aggregates Under Long-Term Fertilization and Its Impact on Organic Carbon Mineralization
		ZHU Yuan-chen¹, ZHANG Min¹, HAN Xiao-zeng¹, LU Xin-chun¹, FENG Hao-liang¹, WU Zhi-min¹, CHEN Xu¹, YAN Jun¹, ZOU Wen-xiu^1*, WANG Wei²
		DOI: 10.3964/j.issn.1000-0593(2025)05-1432-08
		Exploring the contribution of fluorescence characteristics of dissolved organic matter (DOM) in aggregates to soil organic carbon (SOC) mineralization. Based on a 22-year organic fertilizer positioning experiment on black soil, set no fertilization (CK); chemical fertilizer (CF); low amount of organic fertilizer (OM1) added to chemical fertilizer; moderate amount of organic fertilizer (OM2) added to chemical fertilizer; and high amount of organic fertilizer (OM3) added to chemical fertilizer. The composition of soil water-stable aggregates was determined using the wet sieving method, and the spectral characteristics of DOM in the aggregates were investigated using three-dimensional fluorescence spectroscopy coupled with parallel factor (EEM-PARAFAC) technology. At the same time, the mineralization ability of the bulk soil was quantified. Comprehensively revealing how long-term fertilization affects the mineralization of SOC in the whole soil by changing the fluorescence characteristics of aggregate DOM. The results showed that compared with CK, the proportion of particle size aggregates >0.25 mm under the combination of organic fertilizer and chemical fertilizer significantly increased by 4.3%~11.9%. In comparison, the proportion of aggregates with a particle size of <0.053 mm was almost unaffected. EEM-PARAFAC found that combining chemical fertilizers and organic fertilizers can enhance the fluorescence intensity of fulvic-like, protein-like, and humic-like components in various particle-size aggregates of DOM. Among them, humic-like and protein-like components showed the strongest response to organic fertilizers in aggregates of >0.25 mm and <0.25 mm DOM, respectively. Moreover, compared to CK (BIX<0.8), the autotrophic characteristics of DOM in different particle size aggregates were significantly enhanced (BIX>1.0) after the application of chemical fertilizers and organic fertilizers. Still, the fluorescence index (FI) and humification index (HIX) changes were insignificant. In addition, the mineralization ability of SOC increases with the application of organic fertilizers, following the OM3>OM2>OM1>CF>CK pattern. At the end of cultivation (28 d), there was an increase of 68.2%~135.8%. The Mantel test and structural equation (SEM) model indicated that the fluorescence structure changes of aggregates DOM with different particle sizes had an impact on the mineralization of SOC, with the change in BIX value being the main inducing factor, and the protein-like components in DOM mainly controlled the strength of BIX. Furthermore, fertilization mainly drove the mineralization of SOC by improving the particle size distribution of >0.25 mm aggregates. In comparison,<0.25 mm aggregates mainly affected the mineralization of SOC by changing the fluorescence structure of DOM.The research results indicated that the particle size distribution of aggregates and the fluorescence characteristics of DOM, especially the BIX index, can be used to infer the mineralization ability of SOC. This can provide a scientific basis for evaluating and predicting carbon emissions from black soil after long-term application of organic fertilizers.
		2025 Vol. 45 (05): 1432-1439 [Abstract] ( 41 ) RICH HTML PDF (10836 KB) ( 12 )

	1440	Research on the Method of Online Detection of Hollow Watermelons Based on Full-Transmission Near-Infrared Spectroscopy
		LI Jia-qi^{1, 2, 3}, TIAN Xi^{2, 3}, WANG Qing-yan^{2, 3}, HE Xin^{2, 3}, HUANG Wen-qian^{2, 3*}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1440-08
		Watermelon has high nutritional value and is known for its effectiveness in relieving heat in medical applications. Key indicators for evaluating watermelon include ripeness, sweetness, and whether it is hollow. These factors significantly influence market competitiveness. Screening for hollow watermelons ensures higher quality, thereby enhancing market competitiveness. In this study, 307 watermelon spectra were collected using a fully transmissive near-infrared (NIR) spectroscopy device developed independently in our laboratory. Based on the characteristic that hollow areas in watermelons primarily occur at the center of the fruit, we innovatively propose segmenting and weighting the spectra. The optimal two weighted spectra were selected using Support Vector Machine (SVM) and Partial Least Squares Discriminant Analysis (PLSDA) algorithms. Classification models for hollow watermelons were then built using the original spectra and preprocessed with Multiplicative Scatter Correction (MSC) and Savitzky-Golay Smoothing (SGS) in combination with SVM and PLSDA. The results showed that preprocessing the spectra did not necessarily improve the model performance and could even decrease it compared to models built with the original spectra. The models established using the two weighted spectra achieved the best performance, with accuracies of 96.74% (SVM) and 92.39% (PLSDA). The weighted spectra provided better modeling performance than the original and other preprocessed spectra. The weighted spectra were selected using SVM and PLSDA algorithms, and the original spectra were used to establish classification models with one-dimensional convolutional neural networks (1D-CNN). The model accuracies were 98.92% (SVM), 96.77% (PLSDA), and 95.70% (original spectra). The results indicated that 1D-CNN provided better modeling performance than SVM and PLSDA. Additionally, the segmented and weighted spectra remained effective in 1D-CNN and performed better than the original spectra. This study provides important technical support for non-destructive online grading detection of watermelons.
		2025 Vol. 45 (05): 1440-1447 [Abstract] ( 49 ) RICH HTML PDF (16387 KB) ( 39 )

	1448	Identification and Characteristic Analysis of Partial Discharge Emission Spectra of CF₃SO₂F
		DUAN Jun-ran^{1, 2}, GAO Ke-li³, LIU Wei⁴, YAN Xiang-lian³, ZHU Shan⁴, ZHANG Guo-qiang^{1, 2}, HAN Dong^{1, 2}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1448-07
		Trifluoromethylsulfonyl fluoride (CF₃SO₂F) is a novel environmentally friendly insulation gas that has attracted widespread attention due to its excellent insulation strength and environmental friendliness. This paper identifies the emission spectra of CF₃SO₂F under different discharge modes and analyzes the emission spectral characteristics of radiative particles from the discharge perspective. During corona discharge, CF₂ emitted a band spectrum of 200 to 400 nm due to the dissociation of CF₃SO₂F. The associated steady-state decomposition products include CF₄, C₂F₄, C₂F₆, C₃F₈, SO₂F₂, etc. As the energy provided by the electric field increases, overlapping line and band spectra occur in surface discharge emission spectra. A few elemental particles undergo ionization. In spark discharge, where the energy provided by the electric field is highest, line spectra dominate the emission spectra. Multiple levels of ionization of elemental particles occur. With the increase in energy supplied by the electric field, the ratio of photons in the visible and infrared spectral regions exhibits a “red-shift” phenomenon. The research results indicate that the emission spectra formed by gaseous molecules or elemental particles during partial discharge correlate with steady-state decomposition products. The appearance of characteristic spectra confirms the existence of free radical reactions, while the different ratios of photons reflect the amount of energy provided by the electric field. This study can provide references for the basic physical and chemical properties research of CF₃SO₂F.
		2025 Vol. 45 (05): 1448-1454 [Abstract] ( 48 ) RICH HTML PDF (8420 KB) ( 10 )

	1455	Preparation of Ba₉Lu₂Si₆O₂₄:Eu³⁺Red Phosphor and Its Application in White Light-Emitting Diode
		WANG Yun-zheng¹, JI Hong-bo¹, LI Zhao^2, WU Kun-yao^{1, 2}, WANG Ya-nan²
		DOI: 10.3964/j.issn.1000-0593(2025)05-1455-07
		The white light-emitting diode (NUV-WLED) assembled by the near-ultraviolet chip and the red-green-blue phosphor can effectively alleviate the “blue light hazard” and improve the color rendering index. In this study, Ba₉Lu₂Si₆O₂₄:xEu³⁺(0≤x≤0.15) red phosphors were prepared by high-temperature solid-state method, and the phase structure, micromorphology and luminescence properties of the samples were studied by X-ray powder diffraction, scanning electron microscope and spectrometer, and the effect of Eu³⁺ doping on the luminescence properties was analyzed. Eu³⁺ was successfully doped into the Ba₉Lu₂Si₆O₂₄ matrix, and the central excitation peak of the sample was 393 nm (⁷F₀→⁵L₆), the central emission peak was 612 nm (⁵D₀→⁷F₂), and the luminous intensity was the highest when x=0.09. The white LED device was fabricated from Ba₉Lu₂Si₆O₂₄:xEu³⁺ red phosphor, and the color rendering index was close to 90 and showed stable white light emission. The Ba₉Lu₂Si₆O₂₄:xEu³⁺ red phosphor reported in this study has potential application value in white light-emitting diode illumination driven by UV LED chips.
		2025 Vol. 45 (05): 1455-1461 [Abstract] ( 51 ) RICH HTML PDF (22811 KB) ( 16 )

	1462	Preparation and Optical Properties of YVO₄:Yb³⁺/Ho³⁺ Upconversion Luminescent Materials
		JI Hong-bo¹, WANG Yun-zheng¹, LI Zhao^2, WU Kun-yao^{1, 2}, CHEN Wei-xing^1
		DOI: 10.3964/j.issn.1000-0593(2025)05-1462-07
		YVO₄:Yb³⁺/Ho³⁺ up-conversion luminescent materials were synthesized by the high-temperature solid phase method. The samples' phase structure, apparent morphology, and luminescence properties were investigated by X-ray diffractometer, scanning electron microscope, Fourier infrared spectrometer, and fluorescence spectrometer. The results show that the crystal structure of YVO₄ is not changed by doping with different concentrations of rare earth ions, and the phase structure of the sample is pure. The infrared spectrum shows that the sample belongs to the YVO₄ phase of the orthorhombic system. XPS results showed that Yb and Ho were mainly doped into YVO₄ as trivalent ions. SEM analyzed the apparent morphology of the sample to show that the sample is a crystallized and micron-sized powder. The fluorescence spectra showed that the main absorption peaks of YVO₄:Yb³⁺/Ho³⁺ phosphor were 418 nm (⁵I₈→⁵G₅), 455 nm (⁵I₈→⁵G₆), 486 nm (⁵I₈→⁵F₃), 539 nm (⁵I₈→⁵F₄/⁵S₂) and 650 nm (⁵I₈→⁵F₅). The main emission peaks are in the 550 nm (⁵S₂/⁵F₄→⁵I₈) and 660 nm (⁵F₅→⁵I₈) regions. The upconversion luminescence intensity increases first and then decreases with the increase of the Ho³⁺ doping ratio and reaches the highest when the Ho³⁺ doping concentration is 1%. According to Blasse's theory, the ion energy transfer mechanism in YVO₄:Yb³⁺/Ho³⁺ luminescent materials is characterized by multiple dipole moments. The dependence between upconversion luminescence intensity and pump power indicates that red and green light emission belong to the two-photon absorption process. The color coordinates of the samples are all in the red light region, proving that the phosphor is a fluorescent material that can be excited by 980 nm to mainly produce red emission.
		2025 Vol. 45 (05): 1462-1468 [Abstract] ( 44 ) RICH HTML PDF (20382 KB) ( 11 )

	1469	Non-Destructive Detection of Pre-Incubation Breeding Duck Egg Fertilization Information Based on Visible/Near Infrared Spectroscopy and Joint Optimization Strategy
		CHEN Zhuo-ting¹, WANG Qiao-hua^{1, 2*}, WANG Dong-qiao¹, CHEN Yan-bin¹, LI Shi-jun^{1, 2}
		DOI: 10.3964/j.issn.1000-0593(2025)05-1469-07
		The hatching of duck eggs is an important guarantee for producing duck eggs and duck meat. Eggs without sperm cannot hatch ducklings, and they are prone to spoilage in the incubator, which affects the hatching of fertilized eggs. To solve the problems of high labor intensity and resource waste caused by manually removing -sperm-free eggs through egg photography, this paper takes pre-hatching duck eggs as the research object. It proposes a non-destructive detection method for pre-hatching fertilization information of duck eggs based on visible near-infrared spectroscopy and deep learning. This article uses a visible near-infrared fiber optic spectrometer to collect spectral data from 321 Cherry Valley duck eggs (144 fertilized eggs and 177 azoospermia eggs). The spectral data is divided into training and testing sets in a 3∶1 ratio, and the training set is expanded by adding noise and random offset to the original spectral data, randomly selecting and calculating the average spectrum. This article designs an end-to-end deep learning model: the Autoencoder 1DCNN, which uses convolutional and pooling layers instead of the fully connected layers in the autoencoder to obtain an improved convolutional autoencoder CAE. The CAE-1DCNN model is trained using a joint optimization strategy to enable the autoencoder to extract useful features during the data compression reconstruction process and selectively extract features suitable for classification tasks. This article uses three commonly used feature wavelength selection algorithms, namely Competitive Adaptive Reweighted Sampling (CARS), Continuous Projection (SPA), and Uninformative Variable Elimination (UVE), as well as three machine learning classification models, K-Nearest Neighbor (KNN), Naive Bayes (NB), and Random Forest (RF), to combine and compare with the proposed model. The t-distribution Random Neighborhood Embedding (t-SNE) algorithm is used to visualize the feature extraction effect. Finally, this article used a weighted Class Activation Graph (Grad CAM) to visualize the focus areas of spectral data designed in this paper. It explored the biological interpretability of spectral information. The research results indicate that the CAE-1DCNN model proposed in this paper can effectively extract information from spectral data with a discrimination accuracy of 95.06%. The combination of visible near-infrared spectroscopy technology and deep learning can achieve non-destructive detection of pre-incubation fertilization information in duck eggs. The convolutional autoencoder trained using a joint optimization strategy has good feature extraction ability. The end-to-end CAE-1DCNN model facilitates integration and provides technical support for the development of non-destructive testing equipment.
		2025 Vol. 45 (05): 1469-1475 [Abstract] ( 43 ) RICH HTML PDF (6674 KB) ( 15 )

	1476	Hyperspectral Image Detection of Gasoline Pipeline Leakage Using Improved Unet Network
		WANG Ke-ming¹, GONG Wei-jia¹, WANG Hai-ming², CAI Yong-jun², LIU Jia-xing³, SUN Lei⁴, SONG Li-mei¹, LI Jin-yi^1*
		DOI: 10.3964/j.issn.1000-0593(2025)05-1476-09
		In view of the limitations of the low efficiency of gasoline pipeline leak detection and the inability to accurately segment the edge of the leak region, a gasoline pipeline leak detection method is proposed based on hyperspectral image and deep learning. Firstly, the characteristic spectral bands of the two types of gasoline under the background of soil and water were extracted. The continuous projection algorithm was used to reduce the dimensionality of gasoline hyperspectral image data. The gasoline reflectivity was taken as input, and the root-mean-square error was the regression parameter used to obtain 18 characteristic bands near the gasoline reflection peak. Image rotation Angle, horizontal or vertical inversion, and random noise injection into the image are used to expand the dataset sample. Secondly, the Unet hyperspectral image semantic segmentation model is improved, and the network encoder part of Unet is replaced with a dense connection module to strengthen the information exchange between different levels, reduce the computational load, and improve the model detection speed. The spectral attention mechanism module is introduced to make the model pay attention to gasoline image space and spectral features and improve the model detection accuracy. The concept of an inactivation layer is introduced to reduce the complexity of the network by temporarily shutting down some neurons in the network. At the same time, an appropriate time point is set in the training process to implement the early stop strategy to prevent overfitting. Finally, the ablation experiment and comparison experiment were carried out. The results of ablation experiments validate the effectiveness of the dense connection module and the spectral attention mechanism module in improving the network's segmentation accuracy and recall rate. Quantitative comparison experiments on self-built data sets show that the segmentation accuracy of the proposed model for dripping gasoline is 90.34%, and the average detection time of each image is 0.23 s. Compared with Unet, PSE-Unet, and HLCA-Unet models, the average accuracy is increased by 14.39%, 8.01%, and 2.73%, respectively. The recall rate was increased by 8.95%, 8.02%, and 6.55%, and the test time was reduced by 10.83% and 16.97%, respectively, compared with Unet and PSE-Unet models. The qualitative superiority of detection was reflected in the intersection profile of gasoline and background being more consistent with the original image, and the model in this paper could obtain more accurate analysis information of gasoline characteristics. It provides a new technical scheme for gasoline pipeline leakage detection. In addition, compared with the detection of current Unet, PSE-Unet, and HLCA-Unet models on the open Pavia University remote sensing data set, the proposed model still shows better segmentation effect and strong universality and generalization ability; it can be used for many types of hyperspectral image semantic segmentation.
		2025 Vol. 45 (05): 1476-1484 [Abstract] ( 52 ) RICH HTML PDF (24262 KB) ( 13 )

	1485	Classification of Hyperspectral Remote Sensing Images by Joint Hybrid Convolution and Cascaded Group Attention Mechanisms
		WANG Xiao-yan¹, LIANG Wen-hui², BI Chu-ran¹, LI Jie^3*, WANG Xi-yu²
		DOI: 10.3964/j.issn.1000-0593(2025)05-1485-09
		The rich spectral information of hyperspectral remote sensing images can provide reliable data support for their feature classification. However, the problems of high dimensionality and redundancy of spectral data, difficulty associating spatial and spectral features, and insufficient spectral feature extraction have challenged the classification of hyperspectral remote sensing images based on deep learning. Convolutional neural network (CNN) and Vision Transformer (ViT) are two deep learning architectures widely used in computer vision, and each has unique advantages and limitations.CNN is good at capturing local features and spatial hierarchies and can deal with the invariance of the image's translation. ViT can capture global dependencies and has a better understanding of complex patterns in images. To improve the classification accuracy of hyperspectral remote sensing images and give full play to the advantages of both CNN and ViT models, this paper combines the local feature extraction capability of CNN and the global context understanding capability of ViT, and innovatively introduces the 3D Efficient ViT module into the hybrid convolution, and proposes a hyperspectral remote sensing image classification algorithm combining the hybrid convolution and cascading group attention mechanism EVIT3D_HSN: This algorithm introduces 3D Efficient ViT module based on 3D convolution to extract the joint features of hyperspectral remote sensing images and 2D convolution to extract the spatial features, which improves the generalization ability to different datasets and captures the image features of hyperspectral data in a more comprehensive way, thus enhances the performance of the classification algorithm without increasing the complexity of the model. To validate the advancement of this algorithm, this paper's algorithm EVIT3D_HSN is compared with algorithms 1DCNN, 2DCNN, 3DFCN, and 3DCNN and the original algorithm HybridSN for ablation experiments on hyperspectral remote sensing imagery classification datasets India Pines, Pavia University, and Salinas. The classification results of EVIT3D_HSN on the above three datasets are 97.66%, 99.00%, and 99.65% for OA and 97.3%, 98.6%, and 99.6% for the Kappa coefficient, respectively. Compared with 1DCNN, the model classification accuracies are improved by 37.12%, 25.09%, and 33.67%, respectively; compared with 2DCNN, the accuracies are improved by 59%, 57.43%, and 46.92%, respectively; compared with 3DFCN, the accuracies are improved by 45.36%, 24.5% and 29.72%, respectively; and compared with 3DCNN, the accuracies are improved by 28.05%, 14.26% and 34.29%; and compared to HybridSN, the accuracy is improved by 3.76%, 1.85% and 2.57%, respectively. In addition, EVIT3D_HSN has the highest F1 values for a total of 37 features, except stone steel towers for the IP dataset, Painted metal sheets and Shadows for the PU dataset, and Stubble features for the SA dataset. CONCLUSION The experimental results show that EVIT3D_HSN outperforms the above five hyperspectral remote sensing image classification algorithms regarding model accuracy and generalization ability, and the model has good practical value.
		2025 Vol. 45 (05): 1485-1493 [Abstract] ( 51 ) RICH HTML PDF (32081 KB) ( 17 )

	1494	Uneven Distribution and Transformation of Nitrogens During Vitrinite- and Inertinite-Rich Coal Pyrolysis Char of Xiaobaodang Coal Mining Area, Northern Shaanxi Province
		LI Huan-tong¹, ZOU Xiao-yan², ZHANG Ting-ting¹, ZHANG Wei-guo¹, WANG Jun-qi¹
		DOI: 10.3964/j.issn.1000-0593(2025)05-1494-07
		Xiaobaodang raw coal (XR), vitrinite-rich coal (XV) and inertinite-rich coal (XI) were pyrolyzed in a closed high-temperature energy-saving furnace at a heating rate of 5 ℃·min^-1 under high-purity Ar atmosphere. The final pyrolysis temperature was 300~900 ℃. Thermogravimetric-differential thermal analysis (TG-DTA) and X-ray photoelectron spectroscopy (XPS) were used to study the pyrolysis characteristics of raw coal (XR), vitrinite-rich coal (XV) and inertinite-rich coal (XI), and the occurrence of nitrogen in pyrolysis char. The results show that during the pyrolysis process of Xiaobaodang coal, its structure is decomposed and depolymerized to generate and discharge many volatiles. At 440~450 ℃, there is an obvious weight loss peak. The total weight loss rate and maximum weight loss rate of inertinite-rich coal (XI) are smaller than those of vitrinite-rich coal (XV), which is due to the high degree of the aromatization of inertinite and the short length of aromatic ring side chain. The dissociation bond energy of chemical bonds in the structure is large and the thermal stability is high. The main forms of nitrogen in Xiaobaodang coal are pyridine nitrogen (N-6), pyrrole nitrogen (N-5), quaternary nitrogen (N-Q) and nitrogen oxide (N-X). Pyrrole nitrogen (N-5) is higher than pyridine nitrogen (N-6). Pyrrole nitrogen (N-5) in vitrinite-rich coal is slightly lower than in inertinite-rich coal. The quaternary nitrogen (N-Q) content in inertinite-rich coal is slightly higher, mainly due to its high degree of aromatic ring condensation and more nitrogen elements embedded in macromolecular polycyclic aromatic structure. The main forms of nitrogen in pyrolysis char are pyridine nitrogen (N-6) and pyrrole nitrogen (N-5). When the pyrolysis temperature was 300 ℃, the relative content of pyridine nitrogen (N-6) and quaternary nitrogen (N-Q) decreased significantly. With the temperature increase, pyrrole nitrogen (N-5) was converted to pyridine nitrogen (N-6). At 700 ℃, the pyridine-type nitrogen (N-6) in the inertinite-rich coal group has an inflection point, showing a change from rise to fall, possibly caused by aromatic cyclization and polycondensation. At 900 ℃, nitrogen oxides (N-X) disappeared. The surface nitrogen of pyrolysis char is more than that of bulk nitrogen. The removal rates of surface nitrogen in vitrinite-rich coal (XV) and inertinite-rich coal (XI) are 53.49% and 31.86%, respectively. The removal rates of nitrogen in the bulk phase were 33.72% and 15.84%, respectively. At 600 ℃, an inflection point appeared in N/C, showing a change from decrease to increase, which may be caused by the unstable chemical bonds of bridge bonds or alkyl side chains in the structure.
		2025 Vol. 45 (05): 1494-1500 [Abstract] ( 69 ) RICH HTML PDF (5269 KB) ( 27 )

　CopyRight © OFFICE OF SPECTROSCOPY AND SPECTRAL ANALYSIS
　Address: NO.76 Collage South Road of Beijing 　Postalcode:100081
　Tel. (010)62181070　E-mail:chngpxygpfx@vip.sina.com
　Support by Beijing Magtech Co.ltd　E-mail:support@magtech.com.cn