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. 08 Published: 2025-08-01

	2101	Research Progress on the Application of Metasurfaces in Terahertz Super-Resolution Imaging
		LI Guang-ming^{1, 2, 3}, GE Hong-yi^{1, 2, 3}, JIANG Yu-ying^{1, 2, 4}, ZHANG Yuan^{1, 2, 3}, SUN Qing-cheng^{1, 2, 3}, ZHENG Hui-fang^{1, 2, 3}, LI Xing^{1, 2, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2101-09
		Terahertz (THz) radiation, characterized by high penetrability, low energy, and unique fingerprint spectra, has extensive applications in nondestructive testing, biomedical imaging, security screening, and communications. However, traditional THz imaging systems are constrained by the long wavelength of THz waves and the scarcity of natural materials with high transmittance in this spectral range, resulting in limitations in resolution and sensitivity. These shortcomings hinder their ability to meet the technological demands of high-precision and trace-level detection.Metasurfaces, composed of subwavelength-scale structural units, enable precise control of THz wave propagation by modulating electromagnetic waves' phase, amplitude, and polarization. This capability allows them to overcome the diffraction limit of conventional optical systems, offering a viable solution for THz super-resolution imaging. This paper reviews the latest advancements in THz super-resolution imaging using metasurfaces, focusing on various structural types' design principles and application performance. Resonant structures enhance local fields at specific frequencies, enabling high-contrast imaging. Gradient-phase structures guide THz waves with precision through phase gradients. Multilayer metasurfaces leverage stacked-layer designs to achieve complex phase and amplitude modulation. Subwavelength gratings offer superior wavefront control, improving super-resolution imaging. Meanwhile, metamaterial reflective arrays achieve high-resolution wavefront modulation without needing lenses. Key design considerations for high-efficiency THz metasurfaces include structural design strategies, material selection, and precise phase and amplitude modulation techniques. The potential of optimized designs and novel materials in enhancing THz imaging performance and expanding its applications is also analyzed. Future research directions are proposed to address existing challenges in THz metasurface technology, such as complex fabrication processes, limited system compatibility, and material response constraints. These include: (1) developing novel low-loss materials to improve transmission efficiency and phase control; (2) integrating artificial intelligence to optimize metasurface design and performance; and (3) advancing system integration and miniaturization to facilitate the development of portable THz imaging devices for applications in high-precision imaging, medical diagnostics, and security screening. With continuous progress in new materials, intelligent design methodologies, and miniaturization technologies, THz metasurfaces are expected to achieve broader applications. Their advancement will drive the widespread adoption of high-precision imaging and portable devices, fostering new opportunities for scientific discovery and industrial innovation.
		2025 Vol. 45 (08): 2101-2109 [Abstract] ( 27 ) RICH HTML PDF (10543 KB) ( 33 )

	2110	Advances in Laser Ablation Inductively Coupled Plasma Mass Spectrometry for in Situ Microzonation Analysis and Its Application in Forensic Sciencec
		LIU Song, ZHAO Peng-cheng^*, CHEN Wei, YE Zhen, PENG Guo-bin
		DOI: 10.3964/j.issn.1000-0593(2025)08-2110-07
		In recent years, trace physical evidence analysis techniques based on atomic absorption spectroscopy and plasma mass spectrometry have received increasing attention in modern forensic science. Compared with conventional analytical means, atomic absorption spectroscopy and plasma mass spectrometry techniques have the advantages of being fast, sensitive, and simple, but their sensitivity of these techniques is low. Therefore, there is an urgent need for an analytical instrument with high sensitivity, a large range, and the ability to analyze multiple trace elements simultaneously, to meet the requirements for rapid analysis of trace physical evidence in forensic science. Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) is a method that can realize the simultaneous determination of major, trace, and ultra-trace elements. With the advantages of high sensitivity, low detection limit, high spatial resolution, easy operation, and simultaneous determination of multiple elements, this technique has been widely used in in situ microzonation analysis. This paper introduces the advantages of LA-ICP-MS in situ microzonation analysis. For the current status of the application of laser stripping inductively coupled plasma mass spectrometry in forensic science, it reviews the research progress of LA-ICP-MS in situ microzonation analysis technology in the analysis of physical evidence in forensic science, such as glass, hair, inks, gunshot residue (GSR), paper, soil, and so forth, and looks forward to the prospect of the application of this technology.
		2025 Vol. 45 (08): 2110-2116 [Abstract] ( 20 ) RICH HTML PDF (12389 KB) ( 22 )

	2117	Research Progress of Printed Surface-Enhanced Raman Scattering Substrates
		XU Chang, WANG Hui^*, HUANG Bei-qing, LIAN Yu-sheng
		DOI: 10.3964/j.issn.1000-0593(2025)08-2117-07
		Surface-enhanced Raman scattering (SERS) is a spectral analysis method with high sensitivity, fast detection speed, and good accuracy, which can non-destructively detect target substances at extremely low concentrations. Therefore, SERS has important application value in environmental analysis, biomedicine, food safety, and other fields. Due to the properties of additive manufacturing, the application scope of printing technology has expanded from graphic communication to the preparation of functional devices. The substrates used in printing technology have many applications and can easily achieve large-scale fabrication. Combining printing technology with SERS substrate preparation methods is beneficial to developing low-cost batch preparation routes for SERS substrates, thereby promoting the development and application of SERS technology. Therefore, it is significant to study and summarize the printing manufacturing methods of SERS substrates. This review summarizes the progress of research on printed SERS substrates. First, the research progress of several typical SERS-active materials and their applications is introduced, such as noble metals, metal chalcogenides, metal oxides, graphene, and graphene oxide. Then, the preparation methods of printed SERS substrates are summarized. The characteristics of conventional rigid and flexible substrates, such as polymers, paper, and fibers, in printed SERS substrates are analyzed. The principles of printing processes such as gravure printing, screen printing, inkjet printing, micro-contact printing, and 3D printing are discussed, and the application of various substrates and printing processes in the printed preparation of SERS substrates is reviewed. Finally, the future research direction of printed SERS substrates is proposed. Compared with traditional device fabrication methods such as photolithography and etching, printing manufacturing is low-cost and particularly suitable for depositing materials on flexible substrates. It has obvious advantages in the large-scale batch fabrication of flexible SERS substrates. At present, there is plenty of room for the development of printed SERS substrates in both printable materials and printing manufacturing processes. In the future, it is necessary to utilize the advantages of printing technology to develop low-cost and environmentally friendly SERS substrates, and promote the popularization and application of SERS technology.
		2025 Vol. 45 (08): 2117-2123 [Abstract] ( 24 ) RICH HTML PDF (2550 KB) ( 11 )

	2124	Multi Spectral Radiometric Temperature MC-CNN Inversion Algorithm Based on Image Fusion
		XING Jian, ZHU Zi-min, CUI Shuang-long^*
		DOI: 10.3964/j.issn.1000-0593(2025)08-2124-04
		Multispectral radiation thermometry is a method of creating multiple spectral channels in an instrument, using multiple spectral radiance information of the measured target, and processing the data to obtain the temperature. This method has no special requirements for the measured object, making it particularly suitable for simultaneous measurement of temperature and material emissivity of high-temperature targets. Due to the influence of unknown spectral emissivity, the problem of multispectral radiometric temperature inversion can be summarized as an underdetermined equation-solving problem under emissivity constraints. Therefore, multispectral radiometric temperature inversion algorithms have always been a difficult and hot research topic in this field. With the continuous development of deep learning, to fully utilize the precise feature recognition ability of deep learning algorithms in the image field to solve the problem of multispectral radiometric temperature inversion, this paper proposes a multi-channel convolutional neural network (MC-CNN) based on the fusion of Markov Transition Field (MTF) and Gramian Angular Field (GAF) for multispectral radiometric temperature inversion. Deep learning algorithms have obvious advantages in the field of image feature recognition. First, MTF and GAF methods convert one-dimensional spectral voltage data into two-dimensional images with spectral temperature features. Then, the images carrying spectral temperature features are input into an improved convolutional neural network for training, thereby achieving temperature inversion. The simulation results show that for 8 spectral channel data, the average absolute error of inversion at 1K evenly distributed temperature points between 2 355 and 2 624 K is 16.6 K, and the average relative error is 0.7%. Compared with the theoretical value, the inversion error of the measured data of the rocket tail flame is within ±16.5 K, indicating a high inversion accuracy. This method is not affected by unknown emissivity and directly uses spectral voltage data to invert temperature values, further improving the multispectral radiation temperature measurement theory.
		2025 Vol. 45 (08): 2124-2127 [Abstract] ( 19 ) RICH HTML PDF (4947 KB) ( 15 )

	2128	Research on the Chiral Structure of Valine Based on Terahertz Time-Domain Spectral Ellipsometry Technique
		LUO Guo-fang¹, QI Ji², LIU Ya-li², HE Ming-xia², YANG Xin-gang^1, QU Qiu-hong^2, ZHANG Yi-zhu^2*
		DOI: 10.3964/j.issn.1000-0593(2025)08-2128-06
		In this study, we introduce a terahertz (THz) transmission time-domain polarization detection system, along with a calibration method, capable of effectively identifying the optical characteristics of amino acid enantiomers. Although D- and L-valine enantiomers share highly similar physical properties, they exhibit distinct biological activities. The unique spectral properties of THz radiation enable the efficient detection of biomolecular rotational and vibrational modes. By combining THz time-domain spectroscopy with polarizers and a THz broadband phase shifter, we implement THz time-domain spectroscopic ellipsometry (THz-TDSE). This technique allows for the simultaneous measurement of both phase and amplitude of the electric field, facilitating precise polarization control and extraction of relevant ellipsometric parameters.Our findings indicate that the absorption spectra of D- and L-valine are similar in the THz frequency range. However, when modulating the angle of the polarizer and waveplate, a marked difference in absorption coefficients is observed in the high-frequency region. Specifically, at a polarizer angle of -45°, D-valine exhibits a significantly higher absorption coefficient than L-valine, indicating a stronger absorption capability for electromagnetic waves. This research elucidates the spectral response mechanisms of chiral molecules under specific polarization states and demonstrates the potential of THz spectroscopic ellipsometry for chiral substance identification. It paves the way for new applications of THz optical technologies in biochemistry and materials science.
		2025 Vol. 45 (08): 2128-2133 [Abstract] ( 13 ) RICH HTML PDF (4540 KB) ( 9 )

	2134	Wide Range Concentration Measurement of Sulfur Dioxide Based on Adaptive Sliding Window Absorption Spectroscopy
		ZHU Rui¹, DA Yao-dong^{2, 3, 5}, CHANG Yong-qiang^{2, 4, 5}, GAO Jie¹, SHI Teng-da^{2, 3, 5}, ZHANG Yun-gang^1
		DOI: 10.3964/j.issn.1000-0593(2025)08-2134-06
		Sulfur dioxide (SO₂) is an unavoidable pollutant product of the fossil fuel combustion process, and its concentration level is considered a key indicator of combustion efficiency and energy utilization, so it is of great significance to realize the monitoring of SO₂ in the combustion process.Absorption spectroscopy technology has shown a wide range of applications in gas detection due to its high accuracy, strong stability, and non-contact measurement. However, a serious nonlinearity between the absorption intensity and the measured concentration of SO₂ at high concentrations occurs in practical measurements, which interferes with the measurement of SO₂ over a wide range of concentrations. To address this problem, an experimental study was conducted to analyze the treatment of nonlinearity in absorption spectra, and an adaptive sliding window absorption spectroscopy (ASWAS) technique was proposed. In the experiment, the sliding window adaptive traversal and selection of the characteristic absorption band, the first and second screening, and the final inversion were performed to obtain the optimal measurement results. Compared with the wide-band and narrow-band methods, ASWAS shows excellent performance with a relative error of less than 1.3% in the concentration range of 50 to 1 500 ppm.The experimental analysis results show that the measurement system has a measurement accuracy of 0.80% and 0.48% at low and high concentrations, and a stability coefficient of 0.28% and 0.21%, respectively. The above experimental results prove that the proposed measurement system based on ASWAS can realize the measurement of SO₂ concentration in a wide range with high precision and high stability, which provides an effective new idea and method for environmental protection and energy utilization.
		2025 Vol. 45 (08): 2134-2139 [Abstract] ( 20 ) RICH HTML PDF (8966 KB) ( 8 )

	2140	Study on Dual Light Source Measurement of SO₂，NO and NO₂ Based on Differential Optical Absorption Spectroscopy
		DA Yao-dong^{1, 2, 4}, SHI Teng-da^{2, 4}, WANG Zhao², CHANG Yong-qiang^{2, 3, 4}, LI Bing-qian⁵, GAO Jie⁵, ZHANG Yun-gang⁵
		DOI: 10.3964/j.issn.1000-0593(2025)08-2140-09
		The SO₂ and NO_x emitted from boilers significantly impact air quality, and the accurate detection of their emission concentrations is crucial for environmental protection. Ultraviolet differential optical absorption spectroscopy (UV-DOAS) has become an important technique for detecting SO₂ and NO_x emissions due to its advantages, such as high sensitivity and strong anti-interference capability. Normally, portable detection devices of SO₂ and NO_x emissions use deuterium lamps as light sources. However, the mutual interference between NO and NO₂ is unavoidable within the spectral range (180~400 nm), and the detection accuracy is reduced. This paper proposes a dual-light source detection method based on differential absorption spectroscopy of SO₂, NO, and NO₂ concentrations. The high-pressure deuterium lamp and LED are combined to extend the spectral range, and polynomial fitting, wavelet filtering, and least squares method are used to retrieve concentrations from spectral data in the ranges of 295~305 nm (SO₂), 220~230 nm (NO), and 425~450 nm (NO₂). Linear corrections are implemented for high-concentration deviations of SO₂ and NO and low-concentration NO₂ to achieve high-precision quantitative analysis of single-component gases across a wide concentration range. Finally, a mixed gas spectral resolution model was established to address the spectral overlap between SO₂ and NO in the 220~230 nm range, then enables simultaneous concentration detection of mixed gases within ranges of 0～1 000 ppm (SO₂), 0～500 ppm (NO), and 0～500 ppm (NO₂). The lowest detection limits of these three gases were 0.394, 0.30, and 0.78 ppm respectively, with a relative uncertainty of 1.00%.
		2025 Vol. 45 (08): 2140-2148 [Abstract] ( 16 ) RICH HTML PDF (7448 KB) ( 6 )

	2149	Calculation Method of Optical Parameters and Spectral Analysis of Heavy Metals in Water: A Case Study of Typical Lead Compounds
		LIANG Ye-heng^{1, 2, 3}, OUYANG Yu-chun⁴, XU Min-duan⁵, DENG Ru-ru^{1, 2, 3*}, LEI Cong¹, XU Dan⁶, GUO Yu¹, GU Yu-ze¹, LIU Rong¹
		DOI: 10.3964/j.issn.1000-0593(2025)08-2149-07
		Remote sensing-based inversion of heavy metals in water has been a significant challenge within environmental remote sensing monitoring. One important reason is the lack of systematic measurement of optical parameters related to heavy metals. This leads to the absence or inconsistency of input parameters in the inversion models, thereby limiting scholars' ability to conduct in-depth research. In light of this, the study selected four common lead compounds in the effluent of heavy metal pollution enterprises: lead sulfate (PbSO₄), lead tetraoxide (Pb₃O₄), lead chromate (PbCrO₄), and lead sulfide (PbS). We measured their reflectance spectra in the 350~2 500 nm wavelength range and compared their spectral characteristics. Based on this, indoor and outdoor experiments were designed to combine reflectance and transmitted light, proposing a method for calculating the absorption coefficient, scattering coefficient, and extinction coefficient of the samples. The results revealed that the spectral curves of the four lead compounds each exhibited unique features with distinct discriminability. Specifically, the reflectance of PbSO₄ and PbS changed little overall, with PbSO₄ maintaining a reflectance above 80% for most wavelengths, with a variation range of about 17%. PbS remained below 15%, with a variation range of around 7%. The reflectance characteristics of Pb₃O₄ and PbCrO₄ both presented more complex “S” shapes, with multiple characteristic reflection peaks and valleys. At a wavelength of 747 nm as the boundary, the reflectance values relationship among the four compounds showed complex changes before this boundary and stabilized, specifically as PbS443O₄. Furthermore, taking PbSO₄ as an example, the calculated absorption coefficient fluctuated around 0.001 m^-1·mg^-1·L in the 400~900 nm wavelength range, with local minima near 451 and 848 nm, and a local maximum near 725 nm; the scattering coefficient gradually increased from 0.019 to 0.027 m^-1·mg^-1·L, approximately following a line with a very small slope. This indicates that the extinction effect of PbSO₄ on incident light in water is mainly due to scattering, with a minor absorption effect, and that the scattering and extinction coefficients change little with wavelength, similar to “non-selective scattering” in particle scattering types, exhibiting “large particle” scattering characteristics. The above work not only provides a reference for the selection of sensitive bands in the remote sensing inversion process, but also is the basic data necessary to realize the remote sensing inversion of lead concentration in water, provides a methodological reference for the measurement of optical parameters of other heavy metal pollutants, and further lays the foundation for the remote sensing theory of heavy metals in water.
		2025 Vol. 45 (08): 2149-2155 [Abstract] ( 176 ) RICH HTML PDF (4522 KB) ( 43 )

	2156	Dynamical Gemological Characteristics of Irradiated CVD Synthetic Diamond Upon Medium-Low Temperature and Normal Pressure Annealing
		YAN Jun¹, YAN Xue-jun¹, WANG Chao², CHEN Pei-li¹, ZHOU Wu-bang¹, LOU Liu-hong³, ZHENG Tao-jing¹, BAI Hui-min⁴
		DOI: 10.3964/j.issn.1000-0593(2025)08-2156-08
		The optical properties, fluorescence and phosphorescence, and dynamical gemological characteristics of the chemical vapour deposition (CVD) synthetic diamonds subjected to a multistage treatment process, including highpressure high temperature (HPHT) treatment, high energy electron irradiation, and finally medium-low temperature (500～1 000 ℃) and normal pressure annealing, were investigated by DiamondView^TM instrument and photoluminescence (PL) spectrum. The results showed that CVD synthetic diamonds treated with EI could display a series of optical defects, such as 470, 489, 492, 504, 509, 580, 741, and 744 nm. More interestingly, these above defects had different temperature tolerance behaviors. At the annealing temperature of 600 ℃, the characteristic peaks at about 489, 492, and 580 nm disappeared. Moreover, approximately 470, 741, and 744 nm characteristic peaks disappeared at the annealing temperature of 900 ℃. The peak about 504 nmcaused by electron irradiation gradually shifted blue to 503.2 during the annealing temperature rising from 500 to 900 ℃. Meanwhile, the NV vacancies in diamonds recombined or escaped during HPHT treatment. However, these NV defects could occur when the diamond was treated with HPHT, irradiation, and finally annealing. Moreover, the NV concentration increased with the increase in annealing temperature. In addition, the diamond changed from colorless after HPHT treatment to blue-green after heavy EI. It can be finally transformed into orange-yellow or pink with the increasing annealing temperature. The NV vacancy generated during annealing was responsible for the orange, yellow, or pink diamond. The result of this work can provide certain reference for the detection and determination of the synthetic properties of diamonds and their corresponding treatment processes, as well as the development of functional materials in related fields.
		2025 Vol. 45 (08): 2156-2163 [Abstract] ( 17 ) RICH HTML PDF (19908 KB) ( 13 )

	2164	Raman Spectroscopic Detection of the Aging State of Oil-Paper Insulation in Combined Diffusion-Based WGANGP Transformers
		CHEN Xin-gang^{1, 2}, AO Yi¹, ZHANG Zhi-xian^1*, MA Zhi-peng¹, ZHANG Wen-xuan¹, WAN Fu³, KUANG Lu¹, LUO Bo-wen¹
		DOI: 10.3964/j.issn.1000-0593(2025)08-2164-10
		In this paper, a Wasserstein Generative Adversarial Network (WGANGP) method combining Raman spectroscopy with diffusion model improvement is proposed for improving the detection accuracy of the aging state of transformer oil-paper insulation. Raman spectroscopy, with its advantages of no contact and no loss, is used to assess the aging degree of a transformer by analysing the aging products of the oil-paper insulation material inside an oil-immersed power transformer. Combining deep learning classification models simplifies the Raman spectroscopy data preprocessing process, but such models have high requirements on the quantity and quality of training data. The long cycle of thermally accelerated aging experiments results in a relatively scarce set of valid Raman spectral data available for training, limiting the performance of the classification model. To address this challenge, a new data augmentation method, Diffusion-Based WGANGP, is introduced in this study. By combining the forward noise addition process of the denoising diffusion probabilistic model with WGANGP, the method introduces instantiated noise into WGANGP, removes the complex up-sampling process in the generator structure of the traditional WGANGP, simplifies the data augmentation model structure, and facilitates the optimization of model parameters. Compared with the traditional GAN and its variants, this method not only maintains the baseline drift trend of the original eigenpeak features associated with the aging degree in the Raman spectral dataset of the aging samples of the transformer oil-paper insulation, but also maintains an approximate spatial distribution of the features of the original dataset,, and the generated dataset with a Signal-to-Noise Ratio of 24.84 dB, which is improved by 32.11% compared to the original dataset; and at the same time, it also improves the diversity of generated samples, and enhances the generalisation ability, quantitative analysis ability and robustness of the aging diagnosis model based on deep learning. The experimental results show that the Raman spectral dataset generated using the Diffusion-Based WGANGP data augmentation model outperforms other data augmentation methods on several classification models, especially when combined with the ResNet-SVM classification model, in terms of Accuracy (0.997 4), F1 score (0.996 9), Recall (0.996 0) and Precision (0.998 0), which indicates that the improved data augmentation model can effectively solve the problem of the scarcity of samples of transformer aged insulating oil, and at the same time improves the classification model's ability to diagnose the aging state of the transformer quantitatively.
		2025 Vol. 45 (08): 2164-2173 [Abstract] ( 19 ) RICH HTML PDF (10917 KB) ( 12 )

	2174	Reference-Free Reconstruction of Hyperspectral Reflectance Under Complex Illumination Based on Multi-Level Cyclic Optimization Network
		ZHANG He, GAO Kun^*, KE Kun-xin, WANG Jing-yi, ZHANG Ze-feng, HU Bai-yang, YANG Ji-yuan, CHENG Hao-bo
		DOI: 10.3964/j.issn.1000-0593(2025)08-2174-09
		Spectral reflectance reconstruction aims to calculate the true reflectance distribution of an object's surface in response to incident light from hyperspectral images. It is an important research topic in quantitative remote sensing and other hyperspectral imaging applications. Traditional reflectance reconstruction methods mainly rely on statistical models, often requiring stable flat-field illumination conditions or reference images as prior information for reflectance estimation. However, when reference images or other prior information are lacking in practical applications, especially under complex and varying lighting conditions, reconstruction accuracy using traditional methods is not ideal. This paper treats illumination estimation and reflectance reconstruction as a constrained matrix decomposition problem to address this issue. It proposes a multi-level cyclic optimization network model for spectral reflectance reconstruction. The model utilizes a hybrid channel-spatial attention mechanism to adaptively focus on key features in reflectance spectral images, thereby enhancing the extraction and amplification of critical information under non-uniform and multi-illuminant conditions, significantly improving reflectance reconstruction's robustness. Additionally, the network integrates a denoising mechanism comprising two modules: low-rank regularization and total variation regularization. The low-rank regularization module explores the intrinsic low-dimensional structures of illumination and reflectance to suppress noise interference. The total variation regularization module imposes spatial smoothness constraints on the reconstructed spectra, thereby improving reconstruction accuracy, reducing spectral mutations and redundant information, and ensuring spatial consistency throughout the process. To validate the effectiveness of the proposed method, this paper designs related data preprocessing, model training, and evaluation methods. The KAUST hyperspectral dataset is used as the training set in the training process, and different types of incident light source scenarios are simulated in the testing phase. Using the CIE 1964 10° standard observer color matching function as a reference, the hyperspectral images are converted into color images for visualization and quantitative performance evaluation. Experimental results show that the proposed reference-free reflectance reconstruction method outperforms traditional statistical-based reconstruction methods and current popular deep learning-based reconstruction methods regarding reconstruction accuracy indicators such as SAM and GFC. Particularly, without a calibrated whiteboard as a reference, the method still maintains high spectral reconstruction accuracy, demonstrating superior generalization capability and excellent reconstruction performance in complex lighting environments.
		2025 Vol. 45 (08): 2174-2182 [Abstract] ( 17 ) RICH HTML PDF (31522 KB) ( 8 )

	2183	Fast Visual Identification Method of Pyrotechnic Composition Based on Hyperspectral Imaging
		LI Yun-peng, WANG Hong-wei, DAI Xue-jing, WU Lian-quan, HU Wei-cheng, ZHANG Yan-chun
		DOI: 10.3964/j.issn.1000-0593(2025)08-2183-07
		In the explosive scene investigation, the rapid detection and accurate identification of fireworks and explosives play a vital role in preventing, controlling, and rapidly disposing of major explosions. However, the current rapid detection methods for fireworks and explosives mostly have problems such as low recognition speed and difficulty in visualization. Because of this, this paper proposes a method based on hyperspectral imaging technology combined with one-class support vector machine (OCSVM) for rapid detection and recognition of fireworks. Firstly, the hyperspectral data of the sample in the 400~720 nm band were collected with a hyperspectral camera. Principal component analysis (PCA) was used to reduce the dimension of the data, multiplicative scattering correction (MSc) was used to eliminate the baseline offset caused by particle scattering on the sample surface, and Savitzky-Golay (SG) was used to smooth the high-frequency noise and improve the spectral signal-to-noise ratio. Secondly, to reduce the complexity of the model and improve the efficiency, the representative pyrotechnic samples were selected from the spectral data by Kennard stone (K-S) method as the data set, which was divided into the training set and the test set at the ratio of 4∶1. On this basis, the OCSVM model was established. Thirdly, to verify the recognition ability of the model to the pyrotechnic composition, the isolated forest (iforest) and self-encoder (AE) models were established using the same training set, and the recognition ability of the three models to the pyrotechnic composition was compared. Finally, the recognition result is mapped to the RGB image of the test material, and the recognition image is obtained by marking the target pixels with the operation of the mask to realize the visual recognition effect of the pyrotechnic composition. The results show that the overall accuracy of the OCSVM method is higher than 0.95, the F1 score and AUC value are more than 0.8, and the recognition time is less than 2 seconds. The performance of OCSVM in classification accuracy, running speed, F1 score, and AUC is better than that of the isolated forest and self-encoder models. In terms of visual recognition, the recognition image based on the OCSVM model after mapping and mask operation can more accurately reflect the distribution of smoke and powder in all samples. At the same time, the recognition image based on an isolated forest and a self-encoder model can not well reflect the distribution of smoke and powder on yellow paper and black polyester cloth. The research shows that the pyrotechnic identification method based on hyperspectral imaging combined with OCSVM proposed in this paper has the characteristics of high recognition accuracy, fast response speed, and strong generalization ability, and can quickly, accurately, and nondestructively identify pyrotechnics in the test material. Its recognition accuracy, recognition speed-, and visualization effect can be well applied to the rapid discovery and on-site detection of pyrotechnic and explosive at the explosion scene, and provide an effective method for searching for pyrotechnic and explosive in the scene investigation.
		2025 Vol. 45 (08): 2183-2189 [Abstract] ( 18 ) RICH HTML PDF (14430 KB) ( 16 )

	2190	Study on the Interaction Between Lignite Fulvic Acid and Cellulase by Multispectral Method and Molecular Docking Simulation
		WANG Xiao-xia^{1, 2*}, NIU Hao-ran¹, SUN Ji-sheng¹, WANG Ya-xiong^{1, 2}, WANG Meng-han¹, FU Rui^{1, 2}, MAO Qing^{1, 2}, ZHANG Jian-ling³, WANG Wei⁴
		DOI: 10.3964/j.issn.1000-0593(2025)08-2190-10
		In this study, the interaction between fulvic acid (FA) and cellulase (CEL) and its impact on CEL conformation were systematically investigated using fluorescence spectroscopy, synchronous fluorescence spectroscopy, three-dimensional fluorescence spectroscopy, UV-VIS absorption spectroscopy, circular dichroism, infrared spectroscopy, and molecular docking simulations. Fluorescence quenching analysis revealed that the fluorescence intensity of FA decreased in a concentration-dependent manner with increasing CEL concentration. Moreover, the quenching constant decreased with rising temperature, indicating FA's significant static quenching effect on CEL fluorescence. Thermodynamic analysis demonstrated that the interaction between FA and CEL is primarily driven by van der Waals forces and hydrogen bonding, rendering the reaction thermodynamically spontaneous. Förster resonance energy transfer theory suggests potential non-radiative energy transfer between FA and CEL. Ultraviolet-visible absorption and synchronous fluorescence spectra further confirmed the static quenching mechanism and indicated changes in FA's structural density and amino acid residues' microenvironment. Circular dichroism analysis revealed that the binding site of FA and CEL is located in the hydrophobic cavity near the tryptophan residue, leading to reduced hydrophobicity in CEL's secondary structure and an extended peptide chain. This finding corroborates the interaction between FA and CEL and the influence of FA on CEL's secondary structure. Molecular docking studies also showed that the interaction between FA and CEL is predominantly governed by van der Waals forces and hydrogen bonds, achieving stable binding through these interactions, which aligns with the thermodynamic results. Infrared spectroscopy results further support this conclusion.
		2025 Vol. 45 (08): 2190-2199 [Abstract] ( 19 ) RICH HTML PDF (16945 KB) ( 9 )

	2200	*Estimation of Nitrogen Content of Carya illinoinensis* Leaves Based on Canopy Hyperspectral and Wavelet Transform at Different Flight Heights**
		KONG Ling-yuan^{1, 2}, HUANG Qing-feng^{1, 2}, NI Chen^{1, 2}, XU Jia-jia^{1, 2}, TANG Xue-hai^{1, 2*}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2200-10
		Nitrogen is a constituent element of amino acids, proteins, and chlorophyll in plants, which plays an important role in plant photosynthesis. UAV hyperspectral technology can estimate plant nitrogen content non-destructively and efficiently, which is significant for the timely control of tree growth and precise management. Flight height directly affects the accuracy and efficiency of plant information acquisition. In this study, UAV remote sensing images of different resolutions were acquired during the flowering stage of Carya illinoinensis (Changlin and Jiande series) by setting three flight heights (i.e., 40, 60, and 80 m). Thus, the canopy spectra of Carya illinoinensis at the corresponding heights were obtained. Raw hyperspectral data were preprocessed using the continuous wavelet transform (CWT). Furthermore, the response relationship between the LNC of Carya illinoinensis and the canopy spectrum was analyzed by combining two-band spectral indices (i.e., normalized difference spectral index, NDSI). Finally, the competitive adaptive reweighted sampling-iteratively retaining informative variables (CARS-IRIV) algorithm was used to screen the feature variables. Back propagation neural network (BPNN) and random forest (RF) algorithms were used to construct spectral response estimation models for Carya illinoinensis LNC at different heights, to reveal the impact mechanism of UAV flight heights on the canopy spectral characteristics of Carya illinoinensis and LNC. Results showed improved correlation between the canopy spectrum after CWT pretreatment and Carya illinoinensis LNC. CWT combined with NDSI performed better in improving the correlation with LNC. As the flight height increased (from 40, 60 to 80 m), the correlation with the LNC increases for both single-band and two-band spectra.The optimal LNC estimation model was CWT-scale 3-NDSI-BPNN at 40 m flight height, R²_P=0.73, RMSEP=1.13 g·kg^-1, and RPD=1.97. The research results can provide technical support for improving the accuracy of remote sensing estimation of Carya illinoinensis LNC, and further provide a reference for researchers to use a UAV equipped with sensing devices to obtain crop information and set appropriate flight heights.
		2025 Vol. 45 (08): 2200-2209 [Abstract] ( 24 ) RICH HTML PDF (30674 KB) ( 15 )

	2210	A Collaborative Detection Method for Coal Ash Content and Calorific Value Based on A-Unet3+ and Portable NIR Spectrometer
		ZOU Liang¹, KOU Shao-ping¹, REN Ke-long¹, YUAN Guang-fu², XU Zhi-bin³, XU Shi-fan¹, WU Jing-tao^4*
		DOI: 10.3964/j.issn.1000-0593(2025)08-2210-08
		Near-infrared spectroscopy (NIR) technology is increasingly being applied in coal quality analysis due to its rapid and non-destructive advantages. With their compact size and ease of operation, portable NIR analyzers are particularly suited for online coal quality monitoring. However, portable devices typically exhibit lower spectral signal-to-noise ratios than traditional laboratory equipment. Furthermore, existing modeling approaches often focus on individual indicators, overlooking the interdependencies between two coal quality parameters, limiting the models' robustness and prediction accuracy. To address these limitations, this paper proposes a multi-task Unet3+ network model incorporating a channel attention mechanism, aimed at processing the low signal-to-noise ratio and weak-feature spectral data collected by portable NIR analyzers, and achieving the simultaneous prediction of ash content and calorific value in coal. First, a data preprocessing method that combines second-order differentiation with Savitzky-Golay (S-G) convolution smoothing is utilized to effectively reduce noise and enhance spectral peak features, thereby improving the data quality for modeling. Next, a Unet3+ network suitable for spectral data is designed, employing encoders, decoders, and a multi-scale feature fusion module to extract shared features relevant to ash content and calorific value. A channel attention mechanism is introduced to enhance feature representation further. Finally, the model decouples the shared features via fully connected layers to independently learn the specific characteristics of ash content and calorific value, enabling joint prediction for both tasks. Experimental validation on 670 coal samples demonstrates that, compared to typical machine learning methods and conventional deep learning models, the proposed method yields root mean square errors (RMSE) of 2.590 4 and 1.176 3 for ash content and calorific value prediction, respectively. The mean absolute errors (MAE) are 1.964 4 and 0.872 6, with correlation coefficients reaching 0.944 4 and 0.874 3, significantly outperforming the comparison models. These results provide an efficient and accurate solution for the online analysis of coal quality parameters.
		2025 Vol. 45 (08): 2210-2217 [Abstract] ( 14 ) RICH HTML PDF (4956 KB) ( 6 )

	2218	InvertResNet: A Qualitative Analysis Method for Drug Products Based on Deep Learning and Near-Infrared Spectroscopy
		HUANG Tian-yu¹, YANG Hui-hua^{1, 2}, LI Ling-qiao^1*
		DOI: 10.3964/j.issn.1000-0593(2025)08-2218-10
		Classification or qualitative analysis is a key technique for drug traceability and authentication applications. However, in practical applications, we often face technical challenges such as the nonlinear characteristics of near-infrared spectral data, insufficient sample size, data noise interference, and complex preprocessing processes. Traditional machine learning methods cannot fully capture the deep information in spectral data, resulting in limited classification performance. With the rapid development of deep learning technology, its automatic feature extraction and processing capabilities provide a new solution for near-infrared spectral data analysis. In this study, a convolutional neural network named InvertResNet is proposed: the method first converts one-dimensional spectral data into two-dimensional pseudo-images and fills the data with bilinear interpolation during the conversion process to ensure the completeness of the two-dimensionalized spectral data; InvertResNet introduces an inverted residual structure based on the classical convolutional neural network (CNN) framework. By first expanding and then compressing the feature dimensions, the model's depth and width are optimized, effectively suppressing the noise interference and improving the feature extraction and expression capabilities while maintaining the lightweight characteristics. The method adopts a two-dimensional transformation that solves the problem of insufficient data length and preserves the spectral data's local and global spatial correlation, thus enhancing the model's ability to recognize complex patterns and nonlinear information. To evaluate the performance of InvertResNet, this study first utilizes the strawberry puree near-infrared spectral dataset to carry out a preliminary validation of the method, and the results show that it has demonstrated good adaptability and preliminary effectiveness in strawberry puree spectral data processing, which has laid a solid foundation for subsequent in-depth research. Thereafter, the research focus shifts to the publicly available near-infrared spectral classification dataset of pharmaceuticals. On this dataset, the method of this thesis was compared with the traditional partial least squares (PLS), support vector machine (SVM), random forest (RF), standard convolutional neural network (CNN), Swin-Transformer model (SwinTR), GhostNetV2, and SpectraTr model based on the Transformer architecture. Comparative test experiments were conducted. The results show that InvertResNet outperforms traditional algorithms such as PLS and standard CNN structures at different training sample ratios. At low training sample ratios, InvertResNet achieves a classification accuracy of 95.97%, which is significantly better than PLS-DA (79.39%), SVM (68.44%), RF (67.74%), CNN (91.94%), SwinTR (92.74%) and GhostNetV2 (89.91%). With the increase of training samples, the classification accuracy of InvertResNet further improves and reaches 100% under the condition of a high percentage of training samples, which still shows a clear advantage over other models, such as 98.39% for CNN, 98.38% for SwinTR and 98.39% for GhostNetV2. In summary, InvertResNet, with its innovative inverted residual structure and two-dimensional spectral data variation and enhancement method, performs well in the near-infrared spectral analysis of pharmaceuticals, significantly improves the classification accuracy, and has a broad application prospect in the field of near-infrared spectral analysis.
		2025 Vol. 45 (08): 2218-2227 [Abstract] ( 18 ) RICH HTML PDF (14985 KB) ( 9 )

	2228	Construction of a NIR Solid-State Composite Seasoning Freshness AI Model Based on Consumer Sensory Evaluation Ability Assessment
		SHU Qin-da¹, ZHANG Jia-hui^2, WANG Qi³, YUE Bao-hua³, LI Qian-qian^1
		DOI: 10.3964/j.issn.1000-0593(2025)08-2228-06
		To address the issues of intense subjectivity and low reliability in sensory evaluation of umami intensity in solid composite seasonings, this study proposes a prediction model integrating near-infrared spectroscopy (NIRS) and deep learning. By screening 1963 commercial samples and optimizing data quality through consumer sensory evaluation capability assessment, one-dimensional convolutional neural network (1D-CNN) and two-dimensional convolutional neural network (2D-CNN) models were constructed for quantitative prediction. The results showed that without consumer screening, the model achieved a mean relative error (MRE) of 12.79%~15.86% and a correlation coefficient (R) of 0.70~0.74. After excluding data from 6 consumers with poor evaluation capability, the performance of the 2D-CNN model significantly improved (training set: MRE=4.94%, R=0.90; validation set: MRE=5.25%, R=0.87). This study demonstrates that consumer evaluation capability screening and 2D-CNN-based feature extraction effectively enhance prediction accuracy, providing a robust and objective technical solution for quality assessment and product development of solid composite seasonings.
		2025 Vol. 45 (08): 2228-2233 [Abstract] ( 17 ) RICH HTML PDF (5344 KB) ( 12 )

	2234	Determination of Aluminum in Vaccines by Monochromatic Excitation-Energy Dispersive X-Ray Fluorescence Spectrometry
		LIU Cong-cong^{1, 2}, WEN Jia-xin^{1, 3*}, LIN Yuan-heng^{1, 2}, LIANG Wei-yang^{1, 2}, YANG Zhi-ye^{1, 3}, DENG Feng^{1, 3}, LUO Xin-yang⁴, ZHAO Miao-miao⁵
		DOI: 10.3964/j.issn.1000-0593(2025)08-2234-07
		The aluminum adjuvant content in vaccines is an important factor affecting the effectiveness and safety of vaccines. Determining the aluminum content in vaccines swiftly and accurately is an urgent requirement for both production enterprises and inspection institutions. In this paper, the monochromatic excitation technology of a doubly curved crystal was applied to energy dispersive X-ray fluorescence spectrometry, and the matrix effect was corrected by the fundamental parameter method. Samples were pretreated by heating in 30% nitric acid solution at 40 ℃ for 25 minutes. A polypropylene film with a thickness of 4 μm was used as the backing material. Detection was carried out for 300 seconds under the excitation conditions of low voltage and high current, and an analytical method for rapidly determining the aluminum content in vaccines was established. The results showed a significant positive correlation between the measured and theoretical values of aluminum in six vaccine matrices. The correlation coefficients of the fitting equations were all greater than 0.998. The detection limit and quantification limit of the method were 0.027 and 0.090 mg·mL^-1, respectively. Within spiking levels ranging from 0.3 to 0.7 mg·mL^-1, average recovery rates fell between 97.6% and 102%, with precision RSD (n=6) recorded at 2.4% to 6.7%. The coefficient of variation for the same sample measured within 90 days was 4.7%. The accuracy was verified with ICP-OES and ICP-MS as references. The results of 70 batches of vaccine products of six varieties showed good consistency with the reference values, and the absolute values of relative errors did not exceed 10%. This method enables on-site testing without reliance on precision equipment while maintaining accuracy. The research indicates that using monochromatic excitation-energy dispersive X-ray fluorescence spectroscopy offers several advantages, including eliminating the need for standard materials, high accuracy, good precision, and a low detection limit. This technique is suitable for rapid on-site determination of aluminum content in vaccines and holds broad practical significance for widespread application.
		2025 Vol. 45 (08): 2234-2240 [Abstract] ( 19 ) RICH HTML PDF (1660 KB) ( 7 )

	2241	Identification and Raman Spectroscopy Characteristics Analysis of the New Psychoactive Substances Etomidate and Its Analogs
		WANG Shu-dong^1, ZHENG Xuan¹, TIAN Ren-kui² , ZHANG Yan^1, WU Jing-jie^1*
		DOI: 10.3964/j.issn.1000-0593(2025)08-2241-06
		In this study, we report the Raman spectroscopy of etomidate and its analogs medetomidate, propoxate, and the metabolite etomidate acid, and the spectroscopy was analyzed in conjunction with quantum chemical calculations. The results demonstrate distinct Raman spectroscopy differences between etomidate acid and etomidate, due to structural differences, etomidate, medetomidate and propoxate exhibit significantly different Raman activities in the regions of 691～715, 842～866, 955, 1 351～1 375, and 1 411～1 453 cm^-1, which are the key to differentiate the three substances. The investigation further compares the Raman spectroscopy of etomidate and its analogs with those of New Psychoactive Substances(NPS), including piperazine, fentanyl, and cathinone. Potential energy distribution (PED) analysis identifies characteristic Raman peaks near 923, 980, 1 000, 1 030, 1 710, and 1 360 cm^-1 are identified as key indicators for recognizing etomidate and its analogs. Finally, the suspected etomidate “smoke powder” seized by the public security department is tested, and its etomidate content is rapidly confirmed through Raman spectroscopy. The study provides an important reference for the rapid detection of etomidate and its analogs, and also systematically analyzes the differential Raman spectroscopy of etomidate, medetomidate, propoxate and etomidate acid, which makes it possible to differentiate etomidate and its analogs by Raman spectroscopy.
		2025 Vol. 45 (08): 2241-2246 [Abstract] ( 20 ) RICH HTML PDF (3897 KB) ( 10 )

	2247	Diagnostic Method for Brain Glioma Grading Based on Convolutional Neural Networks and Raman Spectroscopy
		XU Qing¹, TANG Jia-wei², LIU Xue-meng³, GUO Jing-xing⁴, ZHU Li-jun¹, ZHOU Qing-qing¹, WANG Liang², LU Guang-ming^1*
		DOI: 10.3964/j.issn.1000-0593(2025)08-2247-06
		Gliomas are the most common primary tumors of the central nervous system, and their pathological grading plays a critical role in guiding treatment decisions and prognostic evaluation. In this study, we retrospectively collected data from 53 patients who underwent glioma surgery at the Eastern Theater General Hospital between January 2023 and January 2024. Among these, 33 cases were high-grade gliomas, and 20 were low-grade gliomas. Raman spectral data of tumor tissue samples were obtained using the InVia laser confocal Raman spectrometer (UK), with 50 points collected for each sample. The spectral data were preprocessed using various methods, including the Savitzky-Golay (SG) algorithm, spectral curve smoothing, and min-max normalization. A convolutional neural network (CNN) was developed to classify gliomas into high- and low-grade categories, and its performance was compared with traditional machine learning models, including support vector machines (SVM), random forests (RF), and decision trees (DT). Each predictive model was evaluated using receiver operating characteristic (ROC) curves, and four key metrics- accuracy, precision, recall, and five-fold cross-validation- were employed to assess model performance. Experimental results demonstrated that the CNN model significantly outperformed the SVM, RF, and DT models in various classification tasks, achieving an area under the curve (AUC) of 0.983 9, compared to 0.915 7 for SVM, 0.903 1 for RF, and 0.780 9 for DT. These findings suggest that integrating Raman spectroscopy with deep learning techniques offers an innovative approach to the grading diagnosis of gliomas. This method improves diagnostic accuracy and efficiency and lays a solid foundation for the future development of automated cancer diagnostic systems.
		2025 Vol. 45 (08): 2247-2252 [Abstract] ( 15 ) RICH HTML PDF (5010 KB) ( 6 )

	2253	*Study on Rapid Antimicrobial Susceptibility Test of Pseudomonas Aeruginosa* by D₂O-Labeled Single-Cell Raman Spectroscopy**
		WANG Feng-chan¹, NIU Lu¹, YE Hai-yan¹, FU Xiao-ting², DAI Jing², LI Yuan-dong², HU Hai-bo¹, LU Xue-chao^1*
		DOI: 10.3964/j.issn.1000-0593(2025)08-2253-06
		Pseudomonas aeruginosa is one of the important pathogens causing clinical pneumonia. The rapid spread of antibiotic resistance threatens our fight against bacterial infections. However, the culture-based broth microdilution method (BMD) is the gold standard method for in vitro antimicrobial susceptibility tests (ASTs), which seriously affects the therapeutic effect of patients due to longer detection time. Single-cell Raman spectroscopy (SCRS) is label-free, culture-free, rapid, accurate and low-cost. Here we research the AST of Pseudomonas aeruginosa using the Clinical Antimicrobials Susceptibility Test Ramanometry (CAST-R), based on D₂O-probed Raman spectroscopy. We selected three antibiotics (Meropenem, Ceftazidime and Cefepime) and three Pseudomonas aeruginosa strains to carry out the AST. CAST-R results show 100% essential agreement and 88.9% categorical agreement with BMD methods, and it can achieve the AST results within 4h. The speed, reliability, and general applicability of CAST-R suggest its potential utility for guiding the clinical administration of antimicrobials.
		2025 Vol. 45 (08): 2253-2258 [Abstract] ( 19 ) RICH HTML PDF (4679 KB) ( 7 )

	2259	An Adaptive Measurement Method for Spectral Lines Based on Local Spectral Trends
		WANG Yu-peng¹, CAI Jiang-hui^2, YANG Hai-feng^2, ZHOU Li-chan¹, SHI Chen-hui¹, LI Yan-feng²
		DOI: 10.3964/j.issn.1000-0593(2025)08-2259-07
		Lick indices are an important metrics for measuring spectral line strength. In the current methods for calculating the Lick index, the wavelength range of the spectral line profile is determined by taking the mean over a fixed interval. This approach often fails to accurately capture the actual strength of individual spectral lines, thereby compromising the accuracy and reliability of the Lick index. To address this problem, this paper introduces an adaptive spectral line measurement method based on local trend characteristics. Firstly, a boundary factor n is defined according to the spectral line characteristics to limit the extreme range of the spectral line profile. Secondly, the slopes on blue and red bands of the core wavelength of the spectral line are calculated to capture the trend information. According to the increasing or decreasing trends changes, n flux peaks (or valleys) closest to the core wavelength on both sides are obtained respectively. Subsequently, the maximum (minimum) flux values and the corresponding wavelengths are selected as the boundary points of the spectral line profile. Finally, based on these boundary points, the line index is calculated through either Equivalent Width (EW) or Magnitude (Mag) formulations, and this value is used to measure the strength of the spectral line feature. By measuring the strength of Ca4227, Hβ, Mgb, and H alpha absorption lines in F, G, and K-type stellar spectra, this paper compares and analyzes the differences between the adaptive and fixed interval measurement methods from three perspectives. First, the scatter distribution map of the Adaptive EW values obtained by the adaptive method and the EW values obtained by the fixed interval method shows that most data points exhibit certain aggregation trends, but are not completely distributed along the diagonal, indicating that both methods are stable. Still, there are systematic differences in the calculation results. Second, the statistical results of two methods show that the mean and standard deviation of Adaptive EW are higher than those of EW, indicating that the adaptive method is more sensitive in capturing spectral detail changes, can capture spectral line changes caused by factors such as stellar individual differences and observational condition fluctuations, and thus more truly reflects the strength of the spectral line. Finally, by visually inspecting the spectral line profile range obtained by the two methods, it is shown that the adaptive method can dynamically adjust the position of the boundary points according to the actual situation of the spectral line, thereby more accurately determining the profile range of the spectral line. Therefore, the adaptive profile boundary detection method establishes an effective approach for quantifying line strengths in individual spectra.
		2025 Vol. 45 (08): 2259-2265 [Abstract] ( 24 ) RICH HTML PDF (10859 KB) ( 17 )

	2266	Eliminating the Influence of Moiré Fringes on Upper Atmospheric Wind Measurement Accuracy in GBAII-DASH
		HUI Ning-ju^{1, 2}, WANG Yan-long¹, LI Wen-wen¹, LIU Yang-he³, LI Cun-xia^{1, 2}, ZHANG Yi-shan¹, FU Di-di¹, TANG Yuan-he^1*
		DOI: 10.3964/j.issn.1000-0593(2025)08-2266-07
		To enhance the detection accuracy of wind speed in the middle and upper atmosphere, GBAII-DASH (Ground-Based Airglow Imaging Interferometer-Doppler Asymmetric Spatial Heterodyne Spectroscopy) imaging interferometer system was developed. The influence of Moiré fringes generated when the periodic interference of straight fringes produced by the system overlapswith the array CCD detector at a certain coupling angle on the wind measurement accuracy of the system was studied, and the wind measurement accuracy of the system was improved. This paper analyzed the cause of Moiré fringes in the GBAII-DASH system. Using the “Four-point algorithm” and the “Fourier transform method”, the forward wind speed was extracted from the interference patterns containing Moiré fringes generated at different coupling angles. The average relative wind measurement errors were 3.07% and 6.89% respectively, so the “Four-point algorithm” had higher wind measurement accuracy. The spectral analysis method simulated the relationship between the difference in spatial frequency between the Moiré fringes and the interference fringes of the GBAII-DASH system and the CCD detector. It was found that increasing the difference could improve the influence of the Moiré fringes on the wind measurement accuracy.The laboratory obtained two types of interference patterns with and without moiré fringes. The wind speed measurement results detected outdoors were compared with the observations of the satellite instrument TIDI (Thermosphere-Ionosphere-Mesosphere Energetics and Dynamics Doppler Interferometer) passing over Xi'an at an altitude of 90~100 km. It was found that the wind speed error of the GBAII-DASH and TIDI in detecting the atmosphere above Xi'an at 90~100 km was only 0.3 m·s^-1. Thus it can be seen that eliminating the moiré fringes and then using the “Four-point algorithm” greatly improves the system's wind measurement accuracy.
		2025 Vol. 45 (08): 2266-2272 [Abstract] ( 14 ) RICH HTML PDF (14462 KB) ( 6 )

	2273	Spectroscopic Character of River Ice Surface and Spectral Analysis of Dissolved Organic Matter in Ice in Inner Mongolia Section of Yellow River
		LENG Yu-peng¹, LI Chun-jiang², YANG Wen-huan², LI Wei-ping², TANG Shi-ke², LI Zhi-jun^1*
		DOI: 10.3964/j.issn.1000-0593(2025)08-2273-08
		The spectral characteristics of the river ice surface are one of the main parameters of remote sensing inversion, which can provide the main basis for studying the optical properties of ice. Ice albe do is an important parameter used to study the energy exchange between the atmosphere and water in cold regions, and is also the most important parameter in spectral characteristics. Dissolved organic matter (DOM) is important in indicating the water environment. Hence, analyzing the spectral characteristics of dissolved organic matter in river ice is of great significance for understanding its environmental indicator function. Using a portable spectrometer to measure the spectra of ice surfaces, the three-dimensional fluorescence spectra of ice samples were combined with parallel factor analysis (PARAFAC). They revealed the spectral characteristics of river ice in the Inner Mongolia section of the Yellow River and the fluorescence components, source characteristics, and influencing factors of DOM, providing a theoretical basis for studying ice's DOM composition and source characteristics. The results indicate that the reflectance of the Yellow River ice increases first and then decreases, and the sediment content significantly impacts the reflectance. One type of protein and one type of terrestrial humus make up most of the ice in the Inner Mongolia portion of the Yellow River; their respective contributions to fluorescence intensity are 56.62% and 43.38%. The ice's surface has the weakest fluorescence, which progressively rises as the thickness of the ice grows. The obtained ice sample's fluorescence index (FI) value is 1.31~2.42, its biological index (BIX) value is 0.74~2.93, and its humification index (HIX) value is 0.20~1.74, according to an analysis of the fluorescence characteristic parameters. Various fluorescence characteristic factors show that endogenous release and external input are the DOM sources of ice samples in the Inner Mongolia part of the Yellow River, with a higher endogenous contribution and a lower degree of humification of ice bodies. This article's research findings provide insight into the spectral properties of river ice in the Yellow River's Inner Mongolia segment, the fluorescence properties of dissolved organic matter, and associated influencing elements, including component origins. In addition to identifying water quality and ecologically monitoring rivers in cold climates, this can offer a theoretical foundation and data support for future studies on the evolution process of DOM in ice.
		2025 Vol. 45 (08): 2273-2280 [Abstract] ( 16 ) RICH HTML PDF (7184 KB) ( 4 )

	2281	Study on Mineralogical and Spectroscopic Characteristics of a New Serpentine From Myanmar
		YU Lian-gang¹, CAI Yi-tao², ZHENG Jin-yu³, LIAO Ren-qing⁴
		DOI: 10.3964/j.issn.1000-0593(2025)08-2281-08
		Aiming at a novel variety of serpentine jade named “Myanmar Lu Yu” with intermingled yellow-green and gray-white colors appearance in the Longling jewellery market of Yunnan, this study employed methods such as petrographic thin-section identification, X-ray powder diffraction, scanning electron microscopy and energy-dispersive spectroscopy, infrared spectroscopy, Raman spectroscopy, and Ultraviolet-visible spectroscopy to explore its mineral composition, chemical components, spectroscopic characteristics, the origin of its captivating color and to infer its ore-forming process. The results show that the jade exhibits fibrous and scaly metamorphic crystalline structures. The gray-white matrix is composed of brucite, while the yellow-green mineral is serpentine, with the serpentine crystals having better orientation than brucite. XRD result shows characteristic diffraction peaks of Antigorite at d₂₀₂=2.525 Å, d_-132=2.618 Å, and d₀₆₀=1.544 Å, and characteristic diffraction peaks at d₁₁₀=4.595 Å and d₀₆₁=1.499 Å of Lizardite. Insights obtained from SEM-EDS elucidate that the FeO_T content in the yellow-green part (5.11%) is much higher than in the gray-white matrix (0.52%). It is speculated that Fe exists in two ways: one is that Fe²⁺ isomorphic substitutes for Mg²⁺ to form ferric Brucite, which is very minor; the other is that Fe²⁺ and Fe³⁺ isomorphic substitutes for Mg²⁺ and Si⁴⁺ enter the crystal structure of serpentine to form ferric serpentine, which constitutes the majority. In addition, there are small amounts of sphalerite, zinc oxide, and trace amounts of metal impurities such as Fe, Co, Ni, Mn, and Cu. Based on the chemical composition and UV-Vis spectroscopic characteristics, it is concluded that the iron element colors the jade. The strong and broad absorption band at 653 nm in the UV-Vis spectrum is caused by Fe²⁺→Fe³⁺ charge transfer, resulting in green color. The moderate absorptions at 435 and 457 nm are due to the spin-forbidden transitions ⁶A₁(⁶S)→⁴_aT₁(4G) and ⁶A₁→⁴E₁+⁴A₁(⁴G) of Fe³⁺, resulting in a yellow color. The superposition of these two absorption effects gives the jade its yellow-green appearance. This jade's uniqueness compared to other serpentine jade varieties lies in the absence of obvious mineral indicators of its genesis. Based on its mineral composition, structure, and color characteristics, it is inferred that this serpentine jade is of ultramafic rock hydrothermal autometasomatic origin, with the formation process divided into three stages: (1) cooling and crystallization differentiation of magnesium-rich ultrabasic magma to form olivineduring its ascent and migration from the mantle; (2) hydrothermal alteration of olivine to serpentine completely during the late stage of magmatic lithogenesis, which is the autometasomatic process of ultrabasic rocks; (3) infiltration of strongly alkaline hot water solutions into fractures after the magmatic period, promoting partial hydrolysis of serpentine to form brucite, which fills the structural fractures in vein-like and patch-like patterns. Serpentine is a product of hydrothermal alteration of ultrabasic magmatic rocks, while brucite is a secondary product formed after the magmatic phase.
		2025 Vol. 45 (08): 2281-2288 [Abstract] ( 21 ) RICH HTML PDF (19221 KB) ( 11 )

	2289	Intelligent Lithology Identification Based on Transfer Learning of Rock Hyperspectral Images
		LI Shan^{1, 2, 3}, LIN Peng^{1, 2, 3}, XU Zhen-hao^{1, 2, 3*}, XIANG Hang^{1, 2, 3}, LI Qian-ji^{1, 2, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2289-13
		The rapid identification of lithology holds significant fundamental geological research significance as well as engineering application value. Traditional lithology recognition primarily depends on the image features of rocks. However, confusion tends to arise when identifying rocks with similar appearances. Consequently, relevant studies further utilize spectral features to reflect the compositional information of rocks. Nevertheless, spectral testing usually demands sample preparation and belongs to the category of destructive testing. This article proposes an intelligent lithology recognition method based on transfer learning of rock hyperspectral images, taking advantage of the integrated imaging hyperspectral technology and the non-destructive, non-contact imaging characteristics. Firstly, the hyperspectral data of the rock region of interest are normalized, and dimensionality reduction is performed to reduce the redundancy of spectral data. Then, a rock hyperspectral image transfer learning model is constructed using a 3D ResNet network, and three-dimensional information is extracted through a residual network. The transfer learning method is reused to train the model by loading pre-trained weights, thereby achieving intelligent recognition of lithology. In this article, the confusion matrix, accuracy (ACC), precision (P), recall (R), and F1 values (F1) are used as evaluation indicators for model accuracy. A comparative analysis is conducted on ResNet101 and ResNet18/34/50 models. The results indicate that the ResNet-101 migration model has the highest accuracy in the test set, reaching 98.29%. The highest P can reach 98.32%, the highest R can reach 98.29%, and the highest F1 can reach 98.31%. The accuracy of ResNet-101 in identifying rock spectral data is over 90% (except for chlorite schist), and most results can even reach 100%. Compared to ResNet18/34/50, ResNet101 has higher recognition accuracy and better stability for identifying each type of rock. In addition, this method was employed to predict the lithology of sampled tunnel site rocks pixel by pixel, verifying the good robustness and generalization performance of the proposed lithology intelligent identification method, which can be used for rapid and intelligent lithology identification in engineering fields like geology, logging, transportation, and water conservancy.
		2025 Vol. 45 (08): 2289-2301 [Abstract] ( 14 ) RICH HTML PDF (36767 KB) ( 7 )

	2302	Early Detection of Northern Corn Leaf Blight Using Hyperspectral Images Combined With One-Dimensional Convolutional Neural Networks
		LU Yang^1, GU Fu-qian¹, GU Ying-nan^2, XU Si-yuan¹, WANG Peng^{3, 4}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2302-09
		Northern corn leaf blight (NCLB) occurs in major maize-producing regions globally, leading to a reduction in both maize quality and yield. Disease identification typically occurs when lesions are more obvious, making it challenging to prevent and control the disease promptly. This study proposes a one-dimensional convolutional neural network (1DCNN) model for early disease detection using hyperspectral imaging. In this research, NCLB was selected as the target disease. After manual inoculation, maize leaves at the silking stage were used for experiments, when lesions had just begun to appear, but the disease could not yet be visually identified. First, hyperspectral images were captured using the SOC710E spectrometer, and spectral data of both healthy and NCLB-infected maize leaves were obtained by selecting regions of interest. Four spectral preprocessing methods Savitzky-Golay smoothing (SG), multiplicative scatter correction (MSC), standard normal variate transformation (SNV), and detrending (DT) were applied to remove noise from the spectral data. Supervised learning algorithms, random forest (RF) and K-nearest neighbors (KNN), were employed for hyperspectral image classification, with accuracy as the evaluation metric. The results indicated that MSC was the optimal preprocessing method, achieving prediction accuracies of 88.13% and 86.26% for the RF and KNN models, respectively. Next, a competitive adaptive reweighted sampling (CARS) algorithm was applied to extract characteristic wavenumbers from the maize leaf spectral data, reducing the original 260 wavenumbers to 48 selected features. Finally, a 1DCNN deep learning model was developed for classification, achieving an accuracy of 99.61%. Compared with traditional classification models such as KNN, RF, partial least squares discriminant analysis (PLS-DA), backpropagation neural network (BP), and support vector machines (SVM), the proposed model improved recognition accuracy by 5.94%, 6.88%, 6.48%, 8.27%, 12.12%, respectively. These findings demonstrate that combining hyperspectral technology with deep learning models provides a new approach and method for early detection of maize diseases, enhancing the accuracy and timeliness of disease recognition.
		2025 Vol. 45 (08): 2302-2310 [Abstract] ( 16 ) RICH HTML PDF (10496 KB) ( 12 )

	2311	Effects of Different Wavelengths of Ultraviolet Light on the Photolysis of Easily Extractable Humus in Severely Rocky Desertification Soils
		KONG Ling¹, WANG Yan^{1, 2, 3}, WANG Ke-qin¹, ZHAO Yang-yi¹, ZHANG Ye-fei^{1, 2, 3*}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2311-06
		This study focuses on the soil of severely rocky desertification,simulating the photolysis process under various wavelengths of ultraviolet(UV) light in drought conditions.The aim is to investigate the decomposition and transformation process of easily extractable humus in severely rocky desertification under varying UV light exposures. The results indicated that: (1) Under various wavelengths of UV irradiation, the total carbon (TC) and total nitrogen (TN) levels in severely rocky desertification soil showed an upward trend as the irradiationtime increased. However, the decrease in the C/N indicated that the irradiation of UV light promoted mineralization and decomposition rate of humus. (2) The contents of primary amine, double bond, carbonate, and O—H of carboxy were stabilized after fluctuating change in the humus easily extracted from the severely rocky desertification soil. They affected the hydrocarbon-generating process to varying degrees depending on various wavelengths of UV irradiation. (3)With increased irradiation time, the substituent group on the benzene ring of the easily extractable humus gradually became dominated by aliphatic chains, the degree of aromatization was enhanced, and the hydrophobic component was increased under UVA and UVB irradiation.In conclusion, ultraviolet radiation in different wavelengths could affect the capacity for hydrocarbon generation, promote the condensation and aromatization of easily extractable humus, and accelerate the process of soil humus mineralization.
		2025 Vol. 45 (08): 2311-2316 [Abstract] ( 16 ) RICH HTML PDF (1955 KB) ( 9 )

	2317	Detecting the Metal Elements and Soil Organic Matter in Farmland by Dual-Modality Spectral Technologies
		WANG Jia-ying¹, ZHU Yu-ting¹, BAI Hao¹, CHEN Ke-ming¹, ZHAO Yan-ru^{1, 2, 3}, WU Ting-ting^{1, 2, 3}, MA Guo-ming⁴, YU Ke-qiang^{1, 2, 3*}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2317-09
		Accurate evaluation of soil quality is one of the prerequisites for ensuring breeding quality, which is of guiding significance for evaluating seed quality and precise fertilization. Soil composition content is an important indicator of soil quality assessment; spectral technology has been proven to detect soil composition quickly and greenly. However, due to the limitations of different spectral excitation principles, single-spectral technology cannot meet the needs of multiple soil composition content detection in breeding fields. This study used laser-induced breakdown spectroscopy (LIBS) and visible-near-infrared spectroscopy (VIS-NIR) combined with intelligent algorithms to analyze 288 soil samples collected from the breeding corn field of Ningxia Runfeng Seed Industry. The prediction models of metal elements and soil organic matter (SOM) content were established, and the spatial visualization distribution of metal elements and SOM content was realized. The specific research is: (1) Detection of metal elements in the maize breeding field. After collecting LIBS spectral data using a collinear double pulse LIBS system, air-PLS was used to correct the baseline of the spectral data and reduce the experimental error. The selected characteristic spectral lines of metal elements were searched in the standard atomic spectrum database of the National Institute of Standards and Technology (NIST). Combined with the LIBS spectrum of national standard soil samples and the true value of metal element content, a partial least squares regression (PLSR) model of four metal elements (Na, K, Mg, Mn) was established in national standard soil samples. Among them, the prediction effect of Mn content was the best, R²_p reached 0.813,RMSEP was 0.155 g·kg^-1; (2) Detection of SOM in maize breeding field. After collecting visible-near infrared spectral data, the spectral data were preprocessed by Savitzky-Golay Convolution Smoothing (SGCS), first derivative transformation, and Multivariate Scattering Correction (MSC), and the PLSR prediction model of SOM content was established to evaluate the three pretreatment methods. The PLSR model established after MSC pretreatment was the best. Subsequently, Monte Carlo cross-validation (MCCV) was used to eliminate the samples of abnormal SOM content. Competitive Adaptive Reweighed Sampling (CARS) and Successive Projections Algorithm (SPA) were used to select the characteristic wavelengths, and the PLSR prediction models of SOM content were established to evaluate the two algorithms. It was concluded that the prediction model's performance, established by the characteristic wavelengths selected by the CARS algorithm, was improved. And the characteristic wavelengths selected by the CARS algorithm and the true value of SOM content were combined to establish PLSR and back propagation neural network (BPNN) prediction models. The PLSR model had the best effect, with R²_p of 0.864, RMSEP of 0.612 g·kg^-1, and RPD_v of 2.733. (3) Spatial visualization distribution of metal elements and SOM content in maize breeding field. The PLSR model established by national standard soil samples was used to predict the content of four metal elements in the maize breeding field, and the spatial distribution map of predicted value content was established. Finally, the SOM content spatial distribution map of the real value, PLSR model predicted value, and BPNN model predicted value was established. The results show that LIBS technology and visible-near infrared spectroscopy quantitative analysis technology can detect the content of metal elements and SOM in the soil of the breeding field, which provides a reference value for the detection and spatial visualization distribution of soil component content.
		2025 Vol. 45 (08): 2317-2325 [Abstract] ( 19 ) RICH HTML PDF (17165 KB) ( 8 )

	2326	Hyperspectral Estimation of Soluble Solids Content in Winter Jujube Based on LSTM-TE Model
		LIU Ao-ran^{1, 2}, MENG Xi², LIU Zhi-guo², SONG Yu-fei^2*, ZHAO Xue-man², ZHI Dan-ning²
		DOI: 10.3964/j.issn.1000-0593(2025)08-2326-09
		Winter jujube is favored by consumers for its rich nutritional content, sweet taste, and excellent flavor. With the increasing demands of consumers for fruit quality, Soluble Solids Content (SSC) has become a key factor in fruit quality evaluation, serving as an important indicator for measuring fruit ripeness and taste quality. Therefore, efficient and non-destructive prediction of winter jujube SSC has significant practical value and importance. This paper proposes an LSTM-TE model, which integrates Long Short-Term Memory (LSTM) networks with a Transformer Encoder, aiming to achieve rapid and non-destructive prediction of winter jujube SSC. By collecting hyperspectral data from 900 winter jujube samples and determining their SSC values, multiple spectral data preprocessing methods (including Multivariate Scatter Correction (MSC), Vector Normalization (VN), Savitzky-Golay (SG) filtering, first derivative (D1), and second derivative (D2)) were applied to process the data. The effects of 10 preprocessing combinations were compared through five models: PLSR, SVR, VGG16, ResNet18, and LSTM, to determine the optimal preprocessing scheme, MSC-SG-D1. Based on this preprocessing method, a multi-model comparison system was further constructed, including PLSR, SVR, VGG16, ResNet18, LSTM, and LSTM-TE, and their performance was analyzed on the test set. Experimental results showed that the LSTM-TE model achieved a coefficient of determination of 0.959 8 and a root mean square error (RMSE) of 1.269 0 on the test set, an improvement of 17.4% compared to the traditional machine learning modelPLSR (R²_p=0.817 3) and 10.9% compared to the single LSTM model (R²_p=0.865 2). This model effectively explored the nonlinear feature relationships in hyperspectral data by capturing temporal features through LSTM and leveraging the global dependency modeling advantages of the Transformer encoder. This study provides a new technical solution for the online detection and grading of winter jujube quality, offering important reference value for applying hyperspectral technology in precision agriculture.
		2025 Vol. 45 (08): 2326-2334 [Abstract] ( 17 ) RICH HTML PDF (21323 KB) ( 7 )

	2335	Combining UAV Digital Imagery With PROSAIL Modeling for LAI Inversion in Summer Maize
		NIU Qing-lin^{1, 4}, ZHANG He-bing^1*, DENG Jiong², FENG Hai-kuan³, LI Chang-chun¹, YANG Gui-jun³, CHEN Zhi-chao¹
		DOI: 10.3964/j.issn.1000-0593(2025)08-2335-13
		The leaf area index (LAI) is an important growth indicator that reflects various maize characteristics and can effectively assist in selecting and breeding new maize varieties. The rapid, non-destructive, and accurate determination of maize LAI is very important in maize breeding. At present, unscrewed aerial vehicle (UAV) visible light remote sensing technology has been rapidly developed in applications for obtaining phenotypic information such as crop LAI because of its advantages in obtaining spatial information about crops in the field in a rapid, non-destructive and high-throughput manner; However, the presence of spectral saturation, due to the lack of information on the response mechanism between spectral parameters and phenotypic information, limits the further improvement of the accuracy of models for estimating phenotypic information; Obviously, the PROSAIL radiative transfer model has the advantage of simulating the response mechanism between crop physicochemical parameters and spectral index parameters, which can effectively enhance the potential of crop physicochemical parameter inversion. Therefore, this study combined UAV digital imagery with the PROSAIL model to invert summer maize LAI to further improve the accuracy of the LAI inversion model. Taking summer maize in the maize breeding experimental field as the research object, a UAV remote sensing system was used to obtain high-resolution digital images at the jointing stage, trumpet stage and tassel emergence stage and combined with the PROSAIL model to construct a summer maize LAI inversion model using partial least squares regression (PLSR), random forest regression (RFR) and convolutional neural network (CNN) regression methods. The results show that (1) based on UAV high-resolution digital images, the model constructed by PLSR regression method has the optimal accuracy, and the R², RMSE and nRMSE of the estimation model and validation model are 0.69, 0.37, 24.28% and 0.73, 0.35, 23.26%, respectively; (2) Based on the PROSAIL model, the model constructed using the RFR regression method has the best accuracy, with R², RMSE and nRMSE of 0.98, 0.28, 6.88% and 0.87, 0.64, 15.97% for the estimated and validated models, respectively; (3) Combining the UAV high-resolution digital imagery with the PROSAIL model, the RFR regression method constructed the model with optimal accuracy, and the R², RMSE and nRMSE of the estimation and validation models were 0.98, 0.27, 7.07% and 0.87, 0.65, 16.35%, respectively. The nRMSE of the optimal estimation model and the validation model were reduced by 17.21% and 6.91%, respectively, compared to using only UAV high-resolution digital imagery. The study shows that combining UAV digital imagery with the PROSAIL model effectively improves the accuracy and stability of the LAI inversion model for summer maize, and provides theoretical guidance to assist in selecting and breeding new maize varieties.
		2025 Vol. 45 (08): 2335-2347 [Abstract] ( 17 ) RICH HTML PDF (12661 KB) ( 6 )

	2348	Estimation of Potato Above-Ground Biomass Using UAV Multispectral Imagery and Biomass Allocation Patterns
		ZHANG Zi-tai^{1, 2}, LI Zhen-hai^1, FAN Yi-guang², MA Yan-peng², GUO Li-xiao², FAN Jie-jie², CHEN Ri-qiang², BIAN Ming-bo², LIU Yang², FENG Hai-kuan^{2, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2348-09
		Above-ground biomass (AGB) is an important indicator for assessing crop growth and guiding agricultural management. Accurately estimating potato AGB is crucial for growth monitoring and yield prediction. Traditional methods of measuring AGB cannot meet the need for large-scale, rapid monitoring. Remote sensing technology is quick, efficient, and non-destructive, making it a valuable tool for crop monitoring. However, remote sensing vegetation indices lose sensitivity in areas with high vegetation coverage, leading to a “saturation effect” that limits AGB accuracy. Most previous studies focused on improving remote sensing indices for AGB estimation. However, these methods still lack clear explanations for how potato growth affects AGB changes. In this study, we analyze the growth process of potatoes and consider the characteristics of different growth stages. We use UAV multispectral data to examine how plant parts, like stems and leaves, contribute to biomass accumulation. We develop a model to track the ratio of leaf biomass (Leaf Ground Biomass, LGB) to AGB throughout the growing season, using effective accumulated temperature (Growing Degree Days, GDD) as an independent variable. By estimating LGB through remote sensing, we can indirectly estimate potato AGB. We built a dynamic AGB allocation model using field measurements. We applied three methods—correlation analysis, variable importance projection (VIP), and random forest feature importance evaluation (FIM)—to select vegetation indices related to LGB. We then used Random Forest (RF), Gaussian Process Regression (GPR), and Partial Least Squares Regression (PLSR) to estimate AGB over the entire growing season. The results showed that: (1) the best vegetation index combination, including MTVI1, MTVI2, NDVI, GNDVI, RVI, and TVI, was selected based on R, VIP, and FIM evaluations; (2) combining the AGB dynamic allocation model with machine learning improved AGB estimation accuracy; (3) the model combining AGB dynamic allocation with random forest achieved the highest accuracy, with a training set R² of 0.80, RMSE of 256.73 kg·ha^-1, and NRMSE of 9.91%. For the validation set, R² was 0.76, RMSE was 211.91 kg·ha^-1, and NRMSE was 11.46%.
		2025 Vol. 45 (08): 2348-2356 [Abstract] ( 17 ) RICH HTML PDF (7165 KB) ( 8 )

	2357	Adsorption Mechanism and Photoelectron Spectroscopy Analysis of Composite Agents on the Surface of Niobium Iron Ore
		HE Yu-long^{1, 3, 4, 5}, HE Xu-ran⁶, ZHAO Zeng-wu^2*, JIA Yan^{1, 3, 4, 5}, CUI Jiu-long^{1, 3, 4, 5}, SUN Pan-shi^{1, 3, 4, 5}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2357-07
		The refractory nature of niobium mineral beneficiation arises from its inherent characteristics of low head grade, finely disseminated grain structure, and complex mineralogical associations. This investigation employs columbite as feed material and introduces a reagent blend as a synergistic collector system for niobium mineral concentration. The interfacial interaction mechanism between the mixed reagent system and columbite particulates was systematically elucidated through Fourier Transform Infrared Spectroscopy (FTIR) and X-ray Photoelectron Spectroscopy (XPS), establishing a technical foundation for comprehensive niobium resource utilization. FTIR characterization revealed chemisorption phenomena through characteristic vibrational signatures: emergence of —N—H stretching vibrations at 3 437.13 cm^-1 with notable peak shifting, coupled with distinct CO (1 651.36 cm^-1) and —C—N (1 102.64 cm^-1) stretching modes. These spectral modifications confirm surface chelation between the reagent blend and columbite lattice constituents. XPS analysis demonstrated surface-specific adsorption evidence through N(1s) photoelectron peak emergence and chemical shift alterations in Nb⁵⁺ binding energy. The observed coordination changes suggest metallo-chelate formation via —O—Nb—O— bonding configurations. Combined spectral interpretations support the formation of five-membered ring chelate complexes at the mineral-reagent interface. Bench-scale flotation tests demonstrated significant metallurgical improvements: The composite collector system elevated niobium concentrate grade from 0.217% to 3.24% with 83.03% recovery. Comparative analysis against single-collector protocols showed equivalent recovery with 0.53% grade enhancement despite yield reduction. This performance optimization confirms the reagent blends' efficacy in enhancing surface hydrophobicity and mineral selectivity, thereby improving overall separation efficiency in niobium flotation circuits.
		2025 Vol. 45 (08): 2357-2363 [Abstract] ( 14 ) RICH HTML PDF (3963 KB) ( 3 )

	2364	Hyperspectral Image Classification of Pigments Based on Multiscale Spatial-Spectral Feature Extraction
		TANG Bin¹, LUO Xi-ling¹, WANG Jian-xu¹, FAN Wen-qi^2, SUN Yu-yu¹, LIU Jia-lu², TANG Huan², ZHAO Ya^3, ZHONG Nian-bing¹
		DOI: 10.3964/j.issn.1000-0593(2025)08-2364-09
		Pigments not only endow cultural relics with color and aesthetic value but also carry rich historical, cultural, and technical information. Accurate classification and identification of pigments are essential for the restoration, preservation, and academic study of ancient painted artworks. Identifying the types and compositions of pigments helps determine the creation period, regional characteristics, and craftsmanship style, providing scientific guidance for restoration and cultural value research. However, traditional pigment analysis faces challenges due to limitations in sample size, surface flatness, and the destructive nature of some analytical methods, which may cause irreversible damage to the artifacts. Hyperspectral Imaging (HSI), with its non-destructive nature, wide-area scanning, and capability of capturing complete spectral information, has become a powerful tool for pigment detection. HSI overcomes the limitations imposed by uneven surfaces and small sample sizes, enabling the extraction of fine-grained spectral and spatial features from pigments. This study aims to utilize HSI for the precise classification and detailed feature extraction of ancient painting pigments, addressing challenges in complex scenarios. We propose a multi-scale spatial-spectral feature fusion method to integrate information at different levels. A spectral-spatial attention mechanism is employed to capture fine details. At the same time, the Vision Transformer (ViT) extracts high-level semantic information from the entire image, enhancing the representation of complex pigment features and improving classification performance. The experimental results show that the proposed method significantly outperforms traditional and other deep learning models in the classification of simulated painting samples: it improves classification accuracy by 34.35% compared to the Support Vector Machine (SVM) and by 8.93% and 5.6% compared to HyBridSN and Spectral-Spatial Residual Network (SSRN), respectively. This study not only improves the accuracy of pigment detection but also provides non-destructive, reliable technical support for the scientific restoration and cultural value preservation of ancient paintings, contributing to the intelligent development of cultural heritage conservation.
		2025 Vol. 45 (08): 2364-2372 [Abstract] ( 13 ) RICH HTML PDF (11897 KB) ( 7 )

	2373	Attenuation Tracking of FY-3B/MERSI Based on Ocean Sun Glint
		HU Xiu-qing^{1, 2, 3}, HE Xing-wei^{1, 2, 3*}, HE Yu-qing⁴, JIANG Meng-die⁴, CHEN Wei⁵
		DOI: 10.3964/j.issn.1000-0593(2025)08-2373-07
		This study focuses on the on-orbit attenuation tracking problem of FY-3B/MERSI. Ocean surface sun glints are adopted as stable targets to track the attenuation from 2011 to 2018. Based on cloud-free and effective sun glint area data, the 865 nm band is regarded as the benchmark band; the ratios between other bands' reflectance and the benchmark band's reflectance are calculated to analyze the attenuations of these bands during the 8 years. There are obvious degradations for all FY-3B/MERSI bands, especially for shortwave bands. The annual degradation rate of 412 nm is 7.12%, while the corresponding value is 0.28% for the 765 nm band. The degradation is much bigger for the 1 030 nm band, at around 3.88%. Furthermore, the reflectance ratios between different bands and benchmark bands show obvious oscillation, consistent with the north-south periodic change of latitude at the center of the sun glint. Ocean surface sun glints are an effective target for the inter-band radiometric calibrations, and could help track the long-term attenuation of on-orbit sensors.
		2025 Vol. 45 (08): 2373-2379 [Abstract] ( 15 ) RICH HTML PDF (7989 KB) ( 4 )

	2380	Judd-Ofelt Analysis and Near Infrared Emissions of Nd³⁺ Doped Oxyfluoride Glass
		FENG Li, YU Yi-huan, DU Hong-li, LIU Chao, CHEN Sai, YANG Chun-cheng
		DOI: 10.3964/j.issn.1000-0593(2025)08-2380-06
		The high-temperature melt quenching method was used to develop Nd³⁺ doped SiO₂-BaF₂-ZnF₂ oxyfluoride glasses. Unlike many literatures that only report the visible absorption of samples, UV-Vis-NIR absorption of the samples was measured in this work. 11 absorption bands were obtained, attributed to transitions of Nd³⁺ from the ground state ⁴I_9/2 to the excited states⁴I_15/2, ⁴F_3/2, ⁴F_5/2+²H_9/2, ⁴F_7/2+⁴S_3/2, ⁴F_9/2, ²H_11/2, ⁴G_5/2+²G_7/2, ⁴G_7/2+²K_13/2+⁴G_9/2, ²G_9/2+²D_3/2+⁴G_11/2+²K_15/2, ²P_1/2+²D_5/2, ²P_3/2+⁴D_5/2+²D_3/2. The Judd analyzed absorption spectra—Ofelt theory, and Judd-Ofelt intensity parameters, spectroscopic quality factor, radiative transition probabilities, fluorescence branching ratios, and radiative lifetimes were obtained.It should be noted that this article presents the radiative transition probabilities, fluorescence branching ratios, and radiative lifetimes of all transitions from the upper to the lower energy level of 11 absorption bands, which is seldom reported in other literature. Near infrared emissions properties of glasses were also investigated, and both SBZ(20) and SBZ(30) exhibit near infrared emissions at 898, 1 059 and 1 328 nm, with the emission at 1 059 nm being significantly stronger than the other two. The values of Δλ_eff of emissions at 1 059 nm for the two samples are the smallest, while the values of σ_emi, σ_emi×Δλ_eff and σ_emi×τ_rad are the largest. The values are: SBZ(20): 36.22 nm, 2.93×10^-20 cm², 10.63×10^-26 cm³, 7.42×10^-24 cm²·s; SBZ(30): 35.42 nm, 2.70×10^-20 cm², 9.58×10^-26 cm³, 7.84×10^-24 cm²·s, which indicate that both samples are potential for laser output.
		2025 Vol. 45 (08): 2380-2385 [Abstract] ( 13 ) RICH HTML PDF (1573 KB) ( 3 )

	2386	Probing Interaction Between Meropenem and NDM-1 by Multispectral Method and Molecular Dynamic Simulation
		LI Jia-chen¹, LI Na¹, LIU Di¹, ZHANG Jia-xin², CHENG Jian-wei³, ZHANG Ye-li^{1, 3*}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2386-07
		The emergence and worldwide spread of metallo-β-lactamase-producing bacteria, particularly New Delhi metallo-β-lactamase (NDM-1), has poseda tremendous challenge in treating drug-resistant bacterial infections. It hydrolyzes almost all β-lactam antimicrobial agents, coupled with the absence of clinically available inhibitors. To comprehend the molecular recognition and interaction between NDM-1 and β-lactam antimicrobial agents, the interaction between NDM-1 and meropenem (MER) was probed by quenching spectroscopy, synchronous fluorescence spectroscopy, circular dichroism spectroscopy, and molecular dynamics simulation. Quenching spectroscopy revealed that MER could cause NDM-1 to undergo endogenous fluorescence quenching and affect the microenvironment of approximately one Trp residue of NDM-1.Synchronous fluorescence spectroscopy displayed that the maximum emission wavelengths of NDM-1 were blue-shifted by 4.0 and 2.0 nm, implying that both Tyr and Trp residues were involved in the binding. Circular dichroism spectroscopy exhibited that the secondary structure of NMD-1 was changed after its interaction with MER, with a decrease in β-sheets, and an increase in irregularly coiled content, suggesting a flexible binding process. In the molecular dynamics results, the β4 (40-47) located near the active pocket of NDM-1 adopted an irregular coil conformation, and loop2 exhibited substantial fluctuations, facilitating NDM-1-MER induced fit. The induced fit effect between NDM-1 and MER was consistent with synchrotron fluorescence and circular dichroism spectroscopy results. Trp93 and His250 amino acid residues formed hydrophobic interactions with MER at the side-chain amino group and the methyl group of the β-lactam ring, respectively, and the amino acid residues of Ile35, Val73, Ala74, Gly36, and Met67 formed van der Waals forces with MER, further promoting the binding. This present study gives crucialinsights into the molecular recognition process of NDM-1 with MER, which may provide new perspectives and a theoretical basis for future development of novel antibiotics and inhibitors targeting this clinically important resistance mechanism.
		2025 Vol. 45 (08): 2386-2392 [Abstract] ( 12 ) RICH HTML PDF (7395 KB) ( 4 )

	2393	Rapid Near-Infrared Detection of Base Baijiu Using Shapley Additive Explanation Algorithm
		ZHANG Gui-yu^{1, 2, 3}, ZHANG Lei^{1, 2, 3}, TUO Xian-guo^{1, 3}, WANG Yi-bo^{1, 3}, XIANG Xing-rui^{1, 3}, YAN Jun^{1, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)08-2393-08
		In current Baijiu extraction processes, the classification of base Baijiu grades is primarily performed using sensory evaluation, and the method is hampered by low detection efficiency and susceptibility to subjective influences. Therefore, near-infrared spectroscopy is applied to base Baijiu grade detection, and the feasibility of using the Shapley additive explanation (SHAP) algorithm from interpretable artificial intelligence for selecting characteristic spectral points is explored. It was found that when the number of features was 36, an accuracy of 97.08% was achieved by the LightGBM predictive model. To further improve model performance, a hybrid strategy combining interval partial least squares (iPLS) with SHAP was proposed, and an accuracy of 99.27% was achieved by the LightGBM model when the number of features was 9. Analysis of the spatial distribution of iPLS interval partitioning and SHAP contribution values indicated that the ranking of SHAP contributions does not strictly correspond to predictive performance. That model's performance can be improved by carefully designing feature selection strategies.
		2025 Vol. 45 (08): 2393-2400 [Abstract] ( 16 ) RICH HTML PDF (5853 KB) ( 9 )

　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