Jianxin Wang , Weiqiang Wang , Jingwei Lv , Famei Wang , Wei Liu , Zao Yi , Qiang Liu , Paul K. Chu , Chao Liu
{"title":"Cantilever-amplified spindle bubble microcavity for high-sensitivity and robust fiber-optic strain sensing","authors":"Jianxin Wang , Weiqiang Wang , Jingwei Lv , Famei Wang , Wei Liu , Zao Yi , Qiang Liu , Paul K. Chu , Chao Liu","doi":"10.1016/j.infrared.2025.106071","DOIUrl":"10.1016/j.infrared.2025.106071","url":null,"abstract":"<div><div>Fiber-optic Fabry-Pérot interferometric (FPI) sensors based on bubble microcavities are fundamentally limited by the sensitivity-robustness trade-off. To overcome this, we propose a spindle-shaped bubble geometry with a cladding-protruding long axis, fabricated via an improved fiber micro-shaping technique using only a commercial fusion splicer. Through parametric optimization guided by experiments and finite element simulations, we demonstrate that the protruding axis acts as a cantilever amplifier, converting axial strain (short-axis direction) into amplified displacement at the long-axis free end, thereby enhancing cavity-length modulation efficiency by 86 %. The optimized structure achieves 49.65 pm/µε strain sensitivity at 1,550 nm while withstanding bending radii ≤ 2.5 cm—surpassing Fully-embedded bubble FPIs by 32.1 % in tensile resistance and 36.2 % in bending tolerance. This innovation bridges the gap between high sensitivity and mechanical robustness, making it ideal for flexible wearables or complex wiring scenarios.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"151 ","pages":"Article 106071"},"PeriodicalIF":3.4,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144828640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Lightweight Target Omni-Directional Enhancement Network for infrared small target detection","authors":"Yichuan Li, Feng He, Qiran Zhang, Wei Zhang","doi":"10.1016/j.infrared.2025.106058","DOIUrl":"10.1016/j.infrared.2025.106058","url":null,"abstract":"<div><div>Due to the limited number of pixels and weak features of small targets in infrared images, detecting such targets in complex backgrounds remains a highly challenging task. It is worthwhile to explore how prior knowledge can be used to compensate for the insufficient inherent information in the original images, thereby assisting deep learning methods in learning more effectively. Inspired by human visual perception, areas with greater local changes tend to attract more attention. In infrared images, while there is some grayscale gradient at the boundary between small targets and the background, background regions also exhibit grayscale variations.To address these issues and make better use of grayscale gradient information as prior knowledge, it is necessary to distinguish the gradients around small targets from those in complex background regions. Therefore, we propose a Target Omnidirectional Enhancement Network (TODENet). The network first uses a Target Enhancement Module to focus on the inherent prior knowledge of infrared images, amplifying the grayscale gradient at the boundary between small targets and the background, while suppressing gradient variations within the background. This approach reduces clutter interference from complex backgrounds and highlights small targets within the image. Building on this, we constructed an Inter-layer Feature Fusion Module based on transposed convolution, which effectively minimizes the loss of high-frequency information of small targets during upsampling. It also makes full use of the semantic information from deep feature maps and the spatial location information from shallow feature maps. Additionally, we developed a Dilated Convolution Module that adjusts the receptive field size to filter out background clutter and then extract fine features of small targets, addressing the problem of losing small target features in the deeper layers of network. Extensive experiments show that TODENet achieves state-of-the-art performance on the NUAA-SIRST, NUDT-SIRST, and IRSTD-1k datasets, with target-level detection rates (Pd) of 97.710%, 99.649%, and 94.218%, respectively. The source code of our work is available at <span><span>https://github.com/LYC-1021/TODE-Net</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"151 ","pages":"Article 106058"},"PeriodicalIF":3.4,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144841336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Pyroelectric characteristics of lead-free materials: A systematic review","authors":"Abhinav Sharma , Sanjay Dhanka , Ankur Kumar , Jasvir Singh Kalsi , Charanjiv Gupta , Ajat Shatru Arora , Surita Maini","doi":"10.1016/j.infrared.2025.106064","DOIUrl":"10.1016/j.infrared.2025.106064","url":null,"abstract":"<div><div>Pyroelectric materials, that generate electrical polarization in response to temperature fluctuations, are widely used in infrared detection, energy harvesting, and biomedical applications. However, environmental and health concerns associated with traditional lead-based materials, such as PZT and PMN-PT, have opened the new window of exploring sustainable lead-free alternatives. This systematic review comprehensively examines recent advancements in lead-free pyroelectric ceramics, focusing on their structural design, performance optimization, and key challenges. Promising materials, including bismuth sodium titanate (BNT), potassium sodium niobate (KNN), and sodium bismuth titanate (NBT)-based ceramics, exhibit competitive pyroelectric coefficients (up to 2720 µC/m<sup>2</sup>K) and thermal stability. Key strategies to enhance performance are; doping, phase boundary engineering, and porosity control, which significantly improve pyroelectric figures of merit (FOMs). For instance, La-doped BNT-BNN ceramics achieve a high pyroelectric coefficient (14.3 × 10<sup>−4</sup>C/m<sup>2</sup>K) with a depolarization temperature of 174 °C, while porous BaTiO<sub>3</sub>-SnO<sub>2</sub> composites demonstrate a 47 % reduction in dielectric constant, enhancing detectivity. Despite these advancements, challenges such as high dielectric loss, thermal instability, and poling inefficiency also persist with these materials. These issues can be addressed through compositional tuning and microstructure optimization, that can enable lead-free materials to surpass conventional lead-based systems. This review provides a roadmap for lead-free pyroelectric technologies, emphasizing the balance between material properties as well as practical feasibility, further focussing on scalable synthesis, thermal endurance, and integration into functional devices to realize their full commercial potential.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"151 ","pages":"Article 106064"},"PeriodicalIF":3.4,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144827031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yangjun Pi , Lingchuan Kong , Bo Yang , Rui Chang , Huayan Pu , Mingliang Zhou , Jun Luo
{"title":"IRTransUNet: Efficient transformer embedding UNet for infrared small target detection","authors":"Yangjun Pi , Lingchuan Kong , Bo Yang , Rui Chang , Huayan Pu , Mingliang Zhou , Jun Luo","doi":"10.1016/j.infrared.2025.106061","DOIUrl":"10.1016/j.infrared.2025.106061","url":null,"abstract":"<div><div>Infrared small target detection is of critical importance in the field of security. However, the inherent weak features and low signal-to-noise ratio of such targets make it particularly difficult to detect them effectively in cluttered and complex backgrounds. To address this issue, this paper proposes IRTransUNet, which integrates local and global information to more thoroughly exploit the differences between the target and the background, thereby achieving more effective discrimination. First, we design a robust feature extractor (RFE), a lightweight and efficient module that leverages a larger contextual receptive field to extract more discriminative fine-grained features. Next, we introduce the IRconvformer module, which focuses on capturing global dependencies and modeling the relationship between the target and background. Specifically, we enhance the target boundary features within tokens using atrous spatial embedding (ASE) and replace the self-attention mechanism with multi-slice linear attention (MSLA), allowing for more efficient global modeling and focused target feature extraction. Additionally, we incorporate a convolutional gated feedforward network (CGFN) to improve the feedforward network, adjusting the information flow between neighboring pixels, thus maintaining the model’s ability to perceive local features. Finally, extensive experiments on four widely used datasets demonstrate that IRTransUNet achieves state-of-the-art performance in infrared small target detection. The code will be publicly available at <span><span>https://github.com/LingchuanK/IRTransUnet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"151 ","pages":"Article 106061"},"PeriodicalIF":3.4,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144860787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shan Tu , Cheng Zhang , Zhongzhou Song , Wentao Zhang , Tao Chen , Yuanpeng Li , Junhui Hu , Heng Xiao , Xianlan Tang , Yanxin Li , Qilin He , Senhao Pang , Jingkai Su
{"title":"High-precision identification of dihydrouracil, glycine anhydride, and piperazine using terahertz spectroscopy","authors":"Shan Tu , Cheng Zhang , Zhongzhou Song , Wentao Zhang , Tao Chen , Yuanpeng Li , Junhui Hu , Heng Xiao , Xianlan Tang , Yanxin Li , Qilin He , Senhao Pang , Jingkai Su","doi":"10.1016/j.infrared.2025.106066","DOIUrl":"10.1016/j.infrared.2025.106066","url":null,"abstract":"<div><div>Distinguishing structurally similar pharmaceutical compounds remains a significant challenge in drug quality assurance, especially when these compounds share overlapping physicochemical properties. This study focuses on three such compounds: dihydrouracil (DHU), glycine anhydride (GA), and piperazine (PIP). DHU and GA are structural isomers, while PIP features a distinct heterocyclic structure, thus providing a rigorous test of the method’s specificity. Traditional analytical techniques, such as high-performance liquid chromatography (HPLC) and nuclear magnetic resonance (NMR), are often limited in rapid on-site deployment due to the need for sample pretreatment and lengthy analysis times. Recent advancements in terahertz time-domain spectroscopy (THz-TDS) have enabled sub-microgram detection and rapid spectral acquisition, making it a promising tool for automated pharmaceutical authentication. In this work, we present a THz-TDS analytical pipeline that leverages machine learning algorithms to differentiate between DHU, GA, and PIP. By employing t-distributed stochastic neighbor embedding (t-SNE) and hierarchical density-based clustering (HDBSCAN) to analyze full-spectrum multivariate patterns, we achieve a clustering accuracy of 99.38 %. This methodology offers a rapid and non-invasive approach to pharmaceutical identification, with significant implications for counterfeit detection, quality assurance, and personalized medicine. It highlights the potential of terahertz (THz) spectroscopy as a transformative tool in modern analytical chemistry.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"151 ","pages":"Article 106066"},"PeriodicalIF":3.4,"publicationDate":"2025-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144841335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Highly sensitive quantitative detection of carbendazim residue based on terahertz metamaterial enhancement and chemometrics","authors":"Jun Hu, Xiaodong Mao, Zhikai Huang, Shimin Yang","doi":"10.1016/j.infrared.2025.106045","DOIUrl":"10.1016/j.infrared.2025.106045","url":null,"abstract":"<div><h3>Objective</h3><div>Carbendazim is widely used as an effective fungicide in agriculture, but its residues on crops pose potential health risks to consumers. This study aims to develop a rapid, non-destructive, and highly sensitive method for detecting carbendazim residues.</div></div><div><h3>Methods</h3><div>Based on electromagnetic theory, this paper presented a terahertz metamaterial sensor incorporating a “cross” compound double-peak structure. Terahertz transmission spectra were collected from 21 different concentration gradients of carbendazim solutions. The spectral response showed a clear decreasing trend in transmission peak amplitude with increasing concentrations. By comparing the results of data preprocessing and feature extraction, the optimal model of terahertz metamaterial detection of carbendazim residue was established.</div></div><div><h3>Result</h3><div>The related coefficient of prediction set (R<sub>P</sub>) and root mean square error of prediction set (RMSEP) of this model are 0.9825 and 0.2001, and the Limit of Detection (LOD) is 0.672 μg/mL.</div></div><div><h3>Conclusion</h3><div>The results demonstrate the feasibility of using terahertz metamaterial sensors combined with spectral analysis for high-sensitivity, non-destructive detection of carbendazim, offering a promising approach for food safety monitoring.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"151 ","pages":"Article 106045"},"PeriodicalIF":3.4,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144826344","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhen Xu , Bohan Liang , Jining Li , Man Luo , Kai Chen , Kai Zhong , Degang Xu
{"title":"6 dB terahertz coupler with Y-shaped plate for free space to transmission waveguide","authors":"Zhen Xu , Bohan Liang , Jining Li , Man Luo , Kai Chen , Kai Zhong , Degang Xu","doi":"10.1016/j.infrared.2025.106043","DOIUrl":"10.1016/j.infrared.2025.106043","url":null,"abstract":"<div><div>This study designs a broadband terahertz coupler by combining a horn-shaped waveguide, a flat waveguide, and a rectangular cavity structure. The influence of different parameters of the structural coupler on the resonant frequency, coupling bandwidth, and coupling efficiency was analyzed. When the distance between the two plates varies within 100 μm, S21 is less than 6 dB in the frequency range of 0.126–0.822 THz, and the coupling bandwidth reaches 0.696 THz. When the length of the rectangular cavity is 100 μm, S21 is less than 5.3 dB in the range of 0.1–1 THz, and it can be used as a strong coupler; When the distance between plates is 900 μm, S21 within the range of 0.1–1 THz is less than 6 dB, which can achieve efficient coupling with an application bandwidth of at least 0.9 THz. Within the frequency range of 0.154–0.533 THz and 0.537–1 THz, both are less than 3 dB. Theoretical analysis and simulation both reveal that as the length of the flat plate increases, the resonant frequency gradually decreases. The maximum coupling efficiency obtained through the transmission terahertz time-domain spectroscopy system test is 76.97 %, which differs from the simulation results by only 6.3 %.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"151 ","pages":"Article 106043"},"PeriodicalIF":3.4,"publicationDate":"2025-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852421","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shichuan Qian , Hui Chao , Zhi Wang , Guanghui Gu , Yinguang Xu , Yulin Wei , Xinping Zhao , Xin Sheng , Jianmei Ren , Shaohua Jin , Lijie Li , Kun Chen
{"title":"Development of monitoring software combined infrared spectra with multivariate curve resolution-alternating least squares (MCR-ALS): Tracking the dynamic synthesis process of 3,7-dinitro-1,3,5,7-tetraazabicyclo[3.3.1]nonane (DPT)","authors":"Shichuan Qian , Hui Chao , Zhi Wang , Guanghui Gu , Yinguang Xu , Yulin Wei , Xinping Zhao , Xin Sheng , Jianmei Ren , Shaohua Jin , Lijie Li , Kun Chen","doi":"10.1016/j.infrared.2025.106041","DOIUrl":"10.1016/j.infrared.2025.106041","url":null,"abstract":"<div><div>The monitoring and analysis of chemical reaction processes are essential for improving reaction efficiency, optimizing production conditions, and ensuring product quality. Infrared spectroscopy, as a non-destructive and real-time analytical technique, provides dynamic insights into the evolution of component concentrations within reaction systems. In this study, the monitoring software was developed by combining infrared spectra with multivariate curve resolution-alternating least squares (MCR-ALS) to dynamically track complex chemical reactions. Using the synthesis of 3,7-dinitro-1,3,5,7-tetraazabicyclo[3.3.1]nonane (DPT) as a case study, the concentration profiles and pure spectra of the reactant hexamethylenetetramine (HMTA), the esterified intermediate (3-acetoxymethyl-7-nitro-1,3,5,7-tetraazabicyclo[3.3.1]nonane), and the product (DPT) were successfully resolved. Computed infrared spectra for HMTA and DPT showed high similarity to measured spectra, with similarity scores of 0.937 and 0.915, respectively. MCR-ALS analysis allowed deduction of the intermediate structure and proposal of the DPT synthesis mechanism, consistent with prior reports and validating the accuracy of the method. Kinetic modeling revealed a three-stage reaction pathway following an A → B → C model, where A, B, and C correspond to HMTA, the esterified intermediate, and DPT, respectively. The process follows first-order kinetics with rate constants of k<sub>1</sub> = 0.095 min<sup>−1</sup> and k<sub>2</sub> = 0.00419 min<sup>−1</sup>. Overall, this study demonstrates the potential of the developed software for monitoring complex reaction systems, supporting process optimization, quality control, and risk management.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"150 ","pages":"Article 106041"},"PeriodicalIF":3.4,"publicationDate":"2025-07-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144721860","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jinyi Li , Yuqing Zhang , Keming Wang , Yue Ji , Zuowei Fu , Limei Song
{"title":"Near infrared laser sensor for multi-species measurement in high temperature flue gas","authors":"Jinyi Li , Yuqing Zhang , Keming Wang , Yue Ji , Zuowei Fu , Limei Song","doi":"10.1016/j.infrared.2025.106040","DOIUrl":"10.1016/j.infrared.2025.106040","url":null,"abstract":"<div><div>A near-infrared (NIR) laser gas analyzer based on tunable diode laser absorption spectroscopy is developed for fast, high-sensitivity and multi-species detection of flue gas components in coal-fired power plant denitrification processes. Two NIR distributed feedback semiconductor lasers are employed to simultaneously cover the absorption lines of ammonia<!--> <!-->(NH<sub>3</sub>), water (H<sub>2</sub>O), and hydrogen chloride (HCl). Wavelength modulation spectroscopy (WMS) scheme is used and distinct sinusoidal modulation amplitudes are applied to selected spectral ranges and optimized accordingly. A dual-channel digital lock-in amplifier circuit is designed to generate driving signals for time-division multiplexed operation of the lasers and to demodulate the detector signal. Both first harmonic (1<em>f</em>) and second harmonic (2<em>f</em>) signals of gas absorption are outputted simultaneously. Gas concentrations are obtained by fitting measured WMS-2<em>f</em>/1<em>f</em> signals with simulations using the Levenberg-Marquardt algorithm. Heat tracing is employed in gas sampling, and gas measurements are carried out in a 5-meter-pathlength gas cell that is operated at an elevated temperature of 523 K. Results demonstrate measurement accuracies of 2.24 % (NH<sub>3</sub>), 2.07 % (HCl), and 0.78 % (H<sub>2</sub>O), measurement sensitivity of 0.0318 ppm, 0.017 ppm, and 0.0104 %, and rise times of 17.4 s, 11.2 s, and 6.4 s, respectively. The performance of our laser sensor demonstrates its potential for applications in multi-species detection of high-temperature flue gas emissions.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"150 ","pages":"Article 106040"},"PeriodicalIF":3.4,"publicationDate":"2025-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144721922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chengyuan Li , Jianwei Ma , Erqiang Zhang , Jinsong Du , Lei Zhang , Min Zhao , Zongying Wang
{"title":"Early detection of mold in cured tobacco leaves based on temporal hyperspectral imaging","authors":"Chengyuan Li , Jianwei Ma , Erqiang Zhang , Jinsong Du , Lei Zhang , Min Zhao , Zongying Wang","doi":"10.1016/j.infrared.2025.106035","DOIUrl":"10.1016/j.infrared.2025.106035","url":null,"abstract":"<div><div>The early detection and warning of mold in tobacco leaves are critical for minimizing losses caused by mold. Existing studies primarily focus on spectral feature analysis at a single time point, overlooking the dynamic evolution of the mold process. To address this limitation, we propose a novel early detection method for mold in tobacco leaves using temporal hyperspectral imaging. First, hyperspectral data of moldy tobacco leaf samples were collected at different time points. Spectral correction and image alignment methods were applied to enhance data quality and ensure spatial consistency across hyperspectral images acquired at different times. To further capture the dynamic characteristics of mold, cumulative energy, and backward short-term energy features are introduced and combined with first- and second-order derivatives, enabling a comprehensive depiction of the temporal behavior of spectral reflectance while effectively identifying local anomalies and long-term trends during the early stages of mold. Additionally, a conditional Wasserstein generative adversarial network with gradient penalty (CWGAN-GP) addresses data imbalance issues, incorporating gradient penalties and conditional information to enhance the quality of generated samples significantly. Experimental results demonstrate that the proposed method effectively detects spectral changes associated with the early stages of mold in tobacco leaves, offering a promising approach for mold warning and real-time monitoring. This study enriches the theoretical framework of hyperspectral image analysis and provides valuable technical support for tobacco quality control.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"150 ","pages":"Article 106035"},"PeriodicalIF":3.1,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144713157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}