Infrared Physics & Technology最新文献_第4页

Ultralow-Noise, widely tunable Stepped-Frequency microwave Carriers Enabled from a Dual-Pump Brillouin cavity 双泵浦布里渊腔实现了超低噪声、宽可调谐的步进频率微波载波

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-19 DOI: 10.1016/j.infrared.2026.106408

Jiaxuan Wang , Yin Xu , Hualong Bao

Stepped-Frequency (SF) signal have been successfully employed in many applications including ranging and vital signs detection. Different application scenarios necessitate SF signal sources with different carrier frequencies, driving a high demand for wideband, reconfigurable, multi-band SF signals that exhibit frequency agility across a broad range. However, generating SF signals that cover the carrier frequency ranges of millimeter waves and terahertz bands using traditional electronic methods poses considerable challenges. Here, we propose and experimentally demonstrate a wide-ranging configurable and ultra-low phase noise carrier source based on dual-pump Brillouin cavity, and generate a SF signal using an optical-recirculating frequency shifting loop, and the SF signal has the characteristics of scanning bandwidth predetermined and flexible frequency-hopping rate. Experimental results show that the proposed system can generate a carrier source with phase noise below −115dBc/Hz at 10 kHz offset at a reconfigurable frequency of 260 to 300 GHz, and can generate a SF signal with a sweep bandwidth of 6.4 GHz in a duration of 14.875 µs. Additionally, the frequency-hopping rate of the SF signal can be adjusted, and an increase rate by a factor of 10–20 is experimentally demonstrated via carefully optimizing the time duration relationships of the applied rectangular optical pulses. This work highlights the potential of the proposed system in generating SF signals with reconfigurable carrier frequencies, sweep bandwidths, and frequency-hopping rates, thereby enhancing the applicability of this system across diverse scenarios.

步进频率（SF）信号已成功地应用于测距和生命体征检测等领域。不同的应用场景需要具有不同载波频率的顺丰信号源，这推动了对宽带、可重构、在宽范围内具有频率敏捷性的多频带顺丰信号的高需求。然而，使用传统的电子方法产生覆盖毫米波和太赫兹频段载波频率范围的SF信号存在相当大的挑战。本文提出并实验验证了一种基于双泵浦布里渊腔的宽范围可配置超低相位噪声载波源，并利用光循环移频环路产生顺频信号，该顺频信号具有扫描带宽预定和跳频灵活的特点。实验结果表明，该系统可以在260 ~ 300 GHz的可重构频率下，在10 kHz偏置下产生相位噪声低于- 115dBc/Hz的载波源，在14.875µs的持续时间内产生扫描带宽为6.4 GHz的顺频信号。此外，SF信号的跳频速率是可以调节的，通过仔细优化所施加的矩形光脉冲的时间持续关系，实验证明了跳频速率增加了10-20倍。这项工作强调了所提出的系统在产生具有可重构载波频率、扫描带宽和跳频速率的顺频信号方面的潜力，从而增强了该系统在不同场景中的适用性。

{"title":"Ultralow-Noise, widely tunable Stepped-Frequency microwave Carriers Enabled from a Dual-Pump Brillouin cavity","authors":"Jiaxuan Wang , Yin Xu , Hualong Bao","doi":"10.1016/j.infrared.2026.106408","DOIUrl":"10.1016/j.infrared.2026.106408","url":null,"abstract":"<div><div>Stepped-Frequency (SF) signal have been successfully employed in many applications including ranging and vital signs detection. Different application scenarios necessitate SF signal sources with different carrier frequencies, driving a high demand for wideband, reconfigurable, multi-band SF signals that exhibit frequency agility across a broad range. However, generating SF signals that cover the carrier frequency ranges of millimeter waves and terahertz bands using traditional electronic methods poses considerable challenges. Here, we propose and experimentally demonstrate a wide-ranging configurable and ultra-low phase noise carrier source based on dual-pump Brillouin cavity, and generate a SF signal using an optical-recirculating frequency shifting loop, and the SF signal has the characteristics of scanning bandwidth predetermined and flexible frequency-hopping rate. Experimental results show that the proposed system can generate a carrier source with phase noise below −115dBc/Hz at 10 kHz offset at a reconfigurable frequency of 260 to 300 GHz, and can generate a SF signal with a sweep bandwidth of 6.4 GHz in a duration of 14.875 µs. Additionally, the frequency-hopping rate of the SF signal can be adjusted, and an increase rate by a factor of 10–20 is experimentally demonstrated via carefully optimizing the time duration relationships of the applied rectangular optical pulses. This work highlights the potential of the proposed system in generating SF signals with reconfigurable carrier frequencies, sweep bandwidths, and frequency-hopping rates, thereby enhancing the applicability of this system across diverse scenarios.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"154 ","pages":"Article 106408"},"PeriodicalIF":3.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023737","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}

引用次数: 0

E2E-AFNet: An End-to-End adaptive NIR-RGB fusion network applied to solid waste recognition 应用于固体废物识别的端到端自适应NIR-RGB融合网络E2E-AFNet

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-19 DOI: 10.1016/j.infrared.2026.106405

Tianchen Ji , Haonan Liu , Jianguo Liu , ChunXiang Liu , FengFeng Jiang , Yibin Xie , Huaiying Fang , Jianhong Yang

Hyperspectral images (HSIs) and RGB multimodal information have proven effective for solid waste recognition. However, optimizing spectral band selection and feature extraction remains a challenge. This paper proposes an End-to-End Adaptive Fusion Network (E2E-AFNet) that integrates Dueling Double Deep Q Network (D3QN) with Near Infrared-RGB (NIR-RGB) feature extraction to achieve unified band selection and feature fusion. Using plastic waste as a case study, we design the Mask-D3QN SBS module to guide spectral input, which is processed by a multispectral feature extraction backbone. This backbone consists of a Multi-Scale Spectral Correlation Unit (MSC Unit) and a Multi-Scale Contour Feature Extraction Unit (MCF Unit), forming a dual-branch structure for feature decoupling. Additionally, the Mutual Attention Feature Interaction Module (MAFIM) efficiently fuses NIR-RGB features for object detection. A reward mechanism based on multimodal detection loss optimizes spectral input selection, enabling end-to-end adaptive fusion. Ablation results show that introducing the MSC and MCF modules improves the F1 score by 6.34 % and 6.45 %, respectively. Their joint use provides an additional ∼ 0.4 % gain, and incorporating the MAFIM module further increases the F1 score by 0.58 %. Further experiments show that the unified band-selection and fusion framework E2E-AFNet outperforms traditional methods, achieving an mAP of 90.48 % and an mAR of 90.87 %. By effectively combining band selection with multi-modal fusion, this approach enhances feature completeness and improves detection performance.

高光谱图像（hsi）和RGB多模态信息已被证明是有效的固体废物识别。然而，优化光谱波段选择和特征提取仍然是一个挑战。提出了一种将Dueling双深Q网络（D3QN）与近红外rgb （NIR-RGB）特征提取相结合的端到端自适应融合网络（E2E-AFNet），实现了统一的频段选择和特征融合。以塑料垃圾为例，设计了Mask-D3QN SBS模块，通过多光谱特征提取主干对输入的光谱进行引导。该主干由一个多尺度光谱相关单元（MSC Unit）和一个多尺度轮廓特征提取单元（MCF Unit）组成，形成双分支结构进行特征解耦。此外，相互关注特征交互模块（meffm）有效融合NIR-RGB特征进行目标检测。基于多模态检测损失的奖励机制优化了频谱输入选择，实现了端到端自适应融合。烧蚀结果表明，引入MSC和MCF模块后，F1分数分别提高了6.34%和6.45%。它们的联合使用提供了额外的~ 0.4%的增益，并且结合mfim模块进一步将F1分数提高了0.58%。进一步的实验表明，统一的频带选择和融合框架E2E-AFNet优于传统的方法，实现了90.48%的mAP和90.87%的mAR。该方法将波段选择与多模态融合有效结合，增强了特征完备性，提高了检测性能。

{"title":"E2E-AFNet: An End-to-End adaptive NIR-RGB fusion network applied to solid waste recognition","authors":"Tianchen Ji , Haonan Liu , Jianguo Liu , ChunXiang Liu , FengFeng Jiang , Yibin Xie , Huaiying Fang , Jianhong Yang","doi":"10.1016/j.infrared.2026.106405","DOIUrl":"10.1016/j.infrared.2026.106405","url":null,"abstract":"<div><div>Hyperspectral images (HSIs) and RGB multimodal information have proven effective for solid waste recognition. However, optimizing spectral band selection and feature extraction remains a challenge. This paper proposes an End-to-End Adaptive Fusion Network (E2E-AFNet) that integrates Dueling Double Deep Q Network (D3QN) with Near Infrared-RGB (NIR-RGB) feature extraction to achieve unified band selection and feature fusion. Using plastic waste as a case study, we design the Mask-D3QN SBS module to guide spectral input, which is processed by a multispectral feature extraction backbone. This backbone consists of a Multi-Scale Spectral Correlation Unit (MSC Unit) and a Multi-Scale Contour Feature Extraction Unit (MCF Unit), forming a dual-branch structure for feature decoupling. Additionally, the Mutual Attention Feature Interaction Module (MAFIM) efficiently fuses NIR-RGB features for object detection. A reward mechanism based on multimodal detection loss optimizes spectral input selection, enabling end-to-end adaptive fusion. Ablation results show that introducing the MSC and MCF modules improves the F1 score by 6.34 % and 6.45 %, respectively. Their joint use provides an additional ∼ 0.4 % gain, and incorporating the MAFIM module further increases the F1 score by 0.58 %. Further experiments show that the unified band-selection and fusion framework E2E-AFNet outperforms traditional methods, achieving an mAP of 90.48 % and an mAR of 90.87 %. By effectively combining band selection with multi-modal fusion, this approach enhances feature completeness and improves detection performance.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"154 ","pages":"Article 106405"},"PeriodicalIF":3.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023685","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}

引用次数: 0

High-efficiency multi-millijoule mid-infrared optical vortex parametric oscillator based on a ZGP crystal 基于ZGP晶体的高效多毫焦耳中红外光学涡旋参量振荡器

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-19 DOI: 10.1016/j.infrared.2026.106413

Disheng Wei , Fei Wang , Lulu Gao , Zhuang Jin , Wenlong Li , Jun Meng , Gaoyou Liu , Zhaojun Liu

In this paper, we demonstrated a high-efficiency, high-energy, and widely tunable mid-infrared optical vortex parametric oscillator based on a ZnGeP₂ (ZGP) crystal, pumped by a 2.05 μm first-order vortex laser operating at a repetition rate of 1 kHz. The oscillator generated a signal beam that consistently carried orbital angular momentum (OAM) across a continuous tuning range from 3.57 to 4.02 μm, while the corresponding idler beam remained OAM-free and exhibited a Gaussian-like intensity profile. At a pump pulse energy of 6.4 mJ, the signal vortex beam achieved a maximum output energy of 2.7 mJ at a central wavelength of 3.86 μm, corresponding to an optical‑to‑optical conversion efficiency (OOCE) exceeding 42 %, which clearly demonstrated efficient OAM transfer from the pump to the signal beam. At maximum output, the signal vortex beam exhibited beam quality factors M² of 2.7 and 2.6 in the x and y directions, respectively.

本文以ZnGeP2 （ZGP）晶体为材料，利用2.05 μm一阶涡旋激光以1khz的重复频率泵浦了一种高效、高能、宽可调谐的中红外光学涡旋参量振荡器。振荡器产生的信号波束在3.57 ~ 4.02 μm的连续调谐范围内始终携带轨道角动量（OAM），而相应的空闲波束则保持无轨道角动量，并表现出类似高斯的强度分布。在泵浦脉冲能量为6.4 mJ时，信号涡旋光束在3.86 μm的中心波长处的最大输出能量为2.7 mJ，对应的光-光转换效率（OOCE）超过42%，这清楚地表明了从泵浦到信号光束的高效OAM传输。在最大输出时，信号涡旋光束在x和y方向上的光束质量因子M2分别为2.7和2.6。

{"title":"High-efficiency multi-millijoule mid-infrared optical vortex parametric oscillator based on a ZGP crystal","authors":"Disheng Wei , Fei Wang , Lulu Gao , Zhuang Jin , Wenlong Li , Jun Meng , Gaoyou Liu , Zhaojun Liu","doi":"10.1016/j.infrared.2026.106413","DOIUrl":"10.1016/j.infrared.2026.106413","url":null,"abstract":"<div><div>In this paper, we demonstrated a high-efficiency, high-energy, and widely tunable mid-infrared optical vortex parametric oscillator based on a ZnGeP<sub>2</sub> (ZGP) crystal, pumped by a 2.05 μm first-order vortex laser operating at a repetition rate of 1kHz. The oscillator generated a signal beam that consistently carried orbital angular momentum (OAM) across a continuous tuning range from 3.57 to 4.02 μm, while the corresponding idler beam remained OAM-free and exhibited a Gaussian-like intensity profile. At a pump pulse energy of 6.4 mJ, the signal vortex beam achieved a maximum output energy of 2.7 mJ at a central wavelength of 3.86 μm, corresponding to an optical‑to‑optical conversion efficiency (OOCE) exceeding 42 %, which clearly demonstrated efficient OAM transfer from the pump to the signal beam. At maximum output, the signal vortex beam exhibited beam quality factors <em>M</em><sup>2</sup> of 2.7 and 2.6 in the <em>x</em> and <em>y</em> directions, respectively.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"154 ","pages":"Article 106413"},"PeriodicalIF":3.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023736","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}

引用次数: 0

Extracavity and intracavity mid-IR lasing in a ZnSe:Fe2+ crystal ZnSe:Fe2+晶体的腔内和腔外中红外激光

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-19 DOI: 10.1016/j.infrared.2026.106414

L. Gafurova, K. Galiuk, A. Sirotkin, B. Ovcharenko, V. Bagdasarov, M. Ponarina, T. Dolmatov, S. Garnov

This paper demonstrates mid-IR lasing in a ZnSe:Fe²⁺ crystal pumped by a train of nanosecond pulses from YLF:Er³⁺/ YSGG:Er³⁺ laser. Two pump configurations are investigated: extra- and intracavity. To the best of our knowledge, this work demonstrates the first implementation of intracavity configurations mid-IR lasing in a ZnSe:Fe²⁺ crystal. The intracavity configuration of the ZnSe:Fe²⁺ laser allows the active crystal, located at Brewster’s angle, to be used both as a gain medium and as a passive shutter for the pump laser, simultaneously. Intracavity and extracavity mid-IR lasing in ZnSe:Fe²⁺ achieve comparable optical-to-optical conversion efficiencies of 3.5 % and 5 %, respectively.

本文演示了YLF:Er3+/ YSGG:Er3+激光器纳秒脉冲泵浦ZnSe:Fe2+晶体的中红外激光。研究了两种泵的配置：腔外泵和腔内泵。据我们所知，这项工作首次在ZnSe:Fe2+晶体中实现了腔内结构的中红外激光。ZnSe:Fe2+激光器的腔内结构允许位于布鲁斯特角的主动晶体同时用作泵浦激光器的增益介质和被动快门。ZnSe:Fe2+的腔内和腔外中红外激光分别实现了3.5%和5%的光-光转换效率。

引用次数: 0

Hierarchical low-rank tensor learning with frequency-domain refinement for infrared small-target detection 基于频域细化的层次低秩张量学习红外小目标检测

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-19 DOI: 10.1016/j.infrared.2026.106406

Meibin Qi , Zijian Zhu , Shuo Zhuang , Liangliang Zhu , Xiaohong Li , Kunyuan Li

Infrared small-target detection (IRSTD) faces fundamental challenges due to low signal-to-clutter ratios and the difficulty of distinguishing faint targets from complex backgrounds and sensor noise. Existing deep learning methods often tackle this as a single problem, lacking a structured approach to disentangle these intertwined components, which limits their performance and interpretability. To address these limitations, we propose HLTL-Net, a novel Hierarchical Low-Rank Tensor Learning Network that enhances target separability through structured decomposition and frequency-domain refinement. Specifically, we introduce a supervised low-rank tensor decomposition (LTD) module that explicitly disentangles an input infrared image into three interpretable subspaces: background, target, and noise. To further enhance the discriminability of the extracted target component, we incorporate a frequency-domain feature refinement (FFR) module that amplifies key frequency signatures while suppressing residual noise. In addition, a multi-stage feature fusion (MFF) module integrates multi-level information from the refined representation to ensure robust localization. Extensive experiments on three public IRSTD benchmarks demonstrate that HLTL-Net achieves state-of-the-art performance, delivering both superior accuracy and improved interpretability for small-target detection. The code is accessible at: https://github.com/master12454/HLTLNet.

红外小目标检测面临着低信杂比、难以从复杂背景和传感器噪声中识别微弱目标的基本挑战。现有的深度学习方法通常将此作为一个单一的问题来解决，缺乏结构化的方法来解开这些纠缠在一起的组件，这限制了它们的性能和可解释性。为了解决这些限制，我们提出了html - net，这是一种新型的分层低秩张量学习网络，通过结构化分解和频域细化来增强目标可分性。具体来说，我们引入了一个监督低秩张量分解（LTD）模块，该模块明确地将输入红外图像分解为三个可解释的子空间：背景、目标和噪声。为了进一步增强提取的目标分量的可分辨性，我们结合了一个频域特征细化（FFR）模块，该模块可以放大关键频率特征，同时抑制残余噪声。此外，多阶段特征融合（MFF）模块集成了来自精炼表示的多层次信息，以确保鲁棒定位。在三个公共IRSTD基准测试上进行的大量实验表明，html - net实现了最先进的性能，为小目标检测提供了卓越的准确性和改进的可解释性。代码可从https://github.com/master12454/HLTLNet访问。

{"title":"Hierarchical low-rank tensor learning with frequency-domain refinement for infrared small-target detection","authors":"Meibin Qi , Zijian Zhu , Shuo Zhuang , Liangliang Zhu , Xiaohong Li , Kunyuan Li","doi":"10.1016/j.infrared.2026.106406","DOIUrl":"10.1016/j.infrared.2026.106406","url":null,"abstract":"<div><div>Infrared small-target detection (IRSTD) faces fundamental challenges due to low signal-to-clutter ratios and the difficulty of distinguishing faint targets from complex backgrounds and sensor noise. Existing deep learning methods often tackle this as a single problem, lacking a structured approach to disentangle these intertwined components, which limits their performance and interpretability. To address these limitations, we propose HLTL-Net, a novel Hierarchical Low-Rank Tensor Learning Network that enhances target separability through structured decomposition and frequency-domain refinement. Specifically, we introduce a supervised low-rank tensor decomposition (LTD) module that explicitly disentangles an input infrared image into three interpretable subspaces: background, target, and noise. To further enhance the discriminability of the extracted target component, we incorporate a frequency-domain feature refinement (FFR) module that amplifies key frequency signatures while suppressing residual noise. In addition, a multi-stage feature fusion (MFF) module integrates multi-level information from the refined representation to ensure robust localization. Extensive experiments on three public IRSTD benchmarks demonstrate that HLTL-Net achieves state-of-the-art performance, delivering both superior accuracy and improved interpretability for small-target detection. The code is accessible at: <span><span>https://github.com/master12454/HLTLNet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"154 ","pages":"Article 106406"},"PeriodicalIF":3.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023688","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}

引用次数: 0

Multifunctional THz chiral metasurfaces for optical dichroism and wavefront shaping 用于光学二色性和波前整形的多功能太赫兹手性超表面

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-18 DOI: 10.1016/j.infrared.2026.106394

Ubaid Ur Rahman Qureshi , Sghaeir Guizani , Faizan Faraz , Zia Ur Rehman , Guizani Sghaier

Chiral terahertz (THz) metasurfaces offer compact platforms for polarization and wavefront control, but most existing designs are passive and limited to a single function. This work proposes a reflective VO₂-based chiral metasurface that combines tunable circular dichroism (CD), linear dichroism (LD), and Pancharatnam–Berry phase wavefront shaping within a single device. The unit cell consists of a gold split-ring resonator partially filled with vanadium dioxide (VO₂), patterned on a polyimide spacer backed by a gold ground plane. The thermally driven insulator–metal transition of VO₂ reconfigures both the amplitude and phase of the reflected THz waves, enabling active control of the device response. Under circularly polarized illumination with VO₂ in the insulating state, the metasurface exhibits CD exceeding 40% from 0.60 to 0.72 THz, with a peak value of 64% at 0.66 THz. For linearly polarized incidence, it maintains LD above 30% from 0.61 to 0.74 THz and reaches a maximum of 54% at 0.69 THz. By rotating the resonator to implement a geometric phase profile, the same platform realizes tunable wavefront manipulation, numerically demonstrating a helicity-dependent anomalous beam deflector, a vortex beam generator carrying orbital angular momentum, and a reflective focusing metalens operating at 0.502 THz and 0.841 THz. These results demonstrate that the proposed VO₂-based chiral metasurface provides a compact and multifunctional solution for thermally reconfigurable polarization control and wavefront shaping in THz imaging and communication systems.

手性太赫兹（THz）超表面为极化和波前控制提供了紧凑的平台，但大多数现有设计都是被动的，仅限于单一功能。本研究提出了一种基于vo2的反射性手性超表面，该超表面在单个器件内结合了可调圆二色性（CD）、线性二色性（LD）和Pancharatnam-Berry相位波前整形。该单元电池由一个部分填充二氧化钒（VO2）的金分裂环谐振器组成，该谐振器被绘制在聚酰亚胺间隔片上，背面是金接平面。VO2的热驱动绝缘体-金属转变重新配置反射太赫兹波的振幅和相位，从而能够主动控制器件响应。在圆偏振光照射下，VO2处于绝缘状态，在0.60 ~ 0.72 THz范围内，超表面CD值超过40%，在0.66 THz范围内CD值最高达64%。对于线极化入射，在0.61 ~ 0.74 THz范围内LD保持在30%以上，在0.69 THz范围内最大LD达到54%。通过旋转谐振器实现几何相位轮廓，同一平台实现了可调谐波前操作，数值演示了一个螺旋相关的异常光束偏转器，一个携带轨道角动量的涡流光束发生器，以及一个工作在0.502 THz和0.841 THz的反射聚焦超透镜。这些结果表明，所提出的基于vo2的手性超表面为太赫兹成像和通信系统中的热可重构偏振控制和波前整形提供了一种紧凑和多功能的解决方案。

{"title":"Multifunctional THz chiral metasurfaces for optical dichroism and wavefront shaping","authors":"Ubaid Ur Rahman Qureshi , Sghaeir Guizani , Faizan Faraz , Zia Ur Rehman , Guizani Sghaier","doi":"10.1016/j.infrared.2026.106394","DOIUrl":"10.1016/j.infrared.2026.106394","url":null,"abstract":"<div><div>Chiral terahertz (THz) metasurfaces offer compact platforms for polarization and wavefront control, but most existing designs are passive and limited to a single function. This work proposes a reflective VO<sub>2</sub>-based chiral metasurface that combines tunable circular dichroism (CD), linear dichroism (LD), and Pancharatnam–Berry phase wavefront shaping within a single device. The unit cell consists of a gold split-ring resonator partially filled with vanadium dioxide (VO<sub>2</sub>), patterned on a polyimide spacer backed by a gold ground plane. The thermally driven insulator–metal transition of VO<sub>2</sub> reconfigures both the amplitude and phase of the reflected THz waves, enabling active control of the device response. Under circularly polarized illumination with VO<sub>2</sub> in the insulating state, the metasurface exhibits CD exceeding 40% from 0.60 to 0.72 THz, with a peak value of 64% at 0.66 THz. For linearly polarized incidence, it maintains LD above 30% from 0.61 to 0.74 THz and reaches a maximum of 54% at 0.69 THz. By rotating the resonator to implement a geometric phase profile, the same platform realizes tunable wavefront manipulation, numerically demonstrating a helicity-dependent anomalous beam deflector, a vortex beam generator carrying orbital angular momentum, and a reflective focusing metalens operating at 0.502 THz and 0.841 THz. These results demonstrate that the proposed VO<sub>2</sub>-based chiral metasurface provides a compact and multifunctional solution for thermally reconfigurable polarization control and wavefront shaping in THz imaging and communication systems.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"154 ","pages":"Article 106394"},"PeriodicalIF":3.4,"publicationDate":"2026-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023687","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}

引用次数: 0

Ranging method of weak lidar signal based on manta ray foraging optimization and variational modal decomposition 基于蝠鲼觅食优化和变分模态分解的弱激光雷达信号测距方法

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-18 DOI: 10.1016/j.infrared.2026.106412

Tianyi Zhang , Xiaobin Xu , Zerun Sun , Shanshan Chen , Zheng Liu , Yingying Ran , Zhiying Tan

Regarding the difficulty of the detection of Lidar weak echo signal, this paper proposes a ranging method based on MRFO algorithm to optimize the VMD parameters and the peak detection. Based on the analysis of Lidar echo signals, MRFO optimization algorithm was used to calculate the optimal VMD decomposition parameters, and the bhattacharyya distance was combined to reconstruct the echo signal. The ranging was then realized by the peak moment identification method. Simulation analysis was conducted to compare the denoising effects with different VMD parameter optimization algorithms. The experimental platform of a Lidar system was built, and it was demonstrated that the laser echo signal completely submerged in noise can be detected by the proposed algorithm. When the ground truths of distances are 46.964 m and 25.363 m, compared with SVD-WT, WT-ST, EMD-ST and DFA-EMD, the de-noising effect and the ranging accuracy of the proposed algorithm are the best. Even though the time cost of proposed algorithm is higher than that of WT-ST, its success rate can reach 89.3 % with different distances, which is more than 20 % better compared with other algorithms. The proposed algorithm provides technical support for faint signal detection for Lidar remote sensing.

针对激光雷达微弱回波信号检测困难的问题，本文提出了一种基于MRFO算法的测距方法来优化VMD参数和峰值检测。在对激光雷达回波信号分析的基础上，采用MRFO优化算法计算最优VMD分解参数，并结合巴塔查里亚距离重构回波信号。然后利用峰矩识别法实现测距。通过仿真分析，比较了不同VMD参数优化算法的去噪效果。搭建了激光雷达系统实验平台，实验结果表明，该算法可以检测到完全淹没在噪声中的激光回波信号。当地面真实距离为46.964 m和25.363 m时，与SVD-WT、WT-ST、EMD-ST和DFA-EMD相比，本文算法的降噪效果和测距精度最好。尽管该算法的时间成本高于WT-ST算法，但在不同距离下，其成功率可达89.3%，比其他算法高出20%以上。该算法为激光雷达遥感微弱信号检测提供了技术支持。

{"title":"Ranging method of weak lidar signal based on manta ray foraging optimization and variational modal decomposition","authors":"Tianyi Zhang , Xiaobin Xu , Zerun Sun , Shanshan Chen , Zheng Liu , Yingying Ran , Zhiying Tan","doi":"10.1016/j.infrared.2026.106412","DOIUrl":"10.1016/j.infrared.2026.106412","url":null,"abstract":"<div><div>Regarding the difficulty of the detection of Lidar weak echo signal, this paper proposes a ranging method based on MRFO algorithm to optimize the VMD parameters and the peak detection. Based on the analysis of Lidar echo signals, MRFO optimization algorithm was used to calculate the optimal VMD decomposition parameters, and the bhattacharyya distance was combined to reconstruct the echo signal. The ranging was then realized by the peak moment identification method. Simulation analysis was conducted to compare the denoising effects with different VMD parameter optimization algorithms. The experimental platform of a Lidar system was built, and it was demonstrated that the laser echo signal completely submerged in noise can be detected by the proposed algorithm. When the ground truths of distances are 46.964 m and 25.363 m, compared with SVD-WT, WT-ST, EMD-ST and DFA-EMD, the de-noising effect and the ranging accuracy of the proposed algorithm are the best. Even though the time cost of proposed algorithm is higher than that of WT-ST, its success rate can reach 89.3 % with different distances, which is more than 20 % better compared with other algorithms. The proposed algorithm provides technical support for faint signal detection for Lidar remote sensing.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"154 ","pages":"Article 106412"},"PeriodicalIF":3.4,"publicationDate":"2026-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023735","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}

引用次数: 0

Improved estimation of soil organic matter content in the Yellow River delta using spectral data optimization and environmental variables integration 利用光谱数据优化和环境变量整合改进黄河三角洲土壤有机质含量估算

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-18 DOI: 10.1016/j.infrared.2026.106409

Shugang Xie , Xueyuan Bai , Cheng Li , Xicun Zhu

Soil Organic Matter (SOM) is a critical component of soil fertility, playing a key role in crop growth and soil improvement. Visible and near-infrared (Vis-NIR) spectroscopy is an ideal method for rapid and accurate estimation of SOM content. This study utilizes fractional-order derivative (FOD) and optimal band combination algorithm (two-dimensional spectral index (2DSI) and three-dimensional spectral index (3DSI)) collaboratively to optimize soil Vis-NIR spectral data and combines environmental variables (EV) to establish an estimation model for SOM content. The results indicate that the use of FOD significantly enhances the correlation between spectra and SOM, especially within the range of 0.4 to 1.4 orders. Modeling outcomes demonstrate that combining FOD with either 2DSI or 3DSI yields superior results compared to models constructed using FOD alone. Specifically, the introduction of 2DSI and 3DSI increases R_v2 values by 3.66 % and 8.54 %, respectively, compared to using FOD alone. Furthermore, when optimized spectral data are combined with EV as input variables, the model’s estimation performance is markedly improved. The optimal model in this study is constructed using a combination of 2Ds + 3Ds + EVs, achieving R_v², RMSE_v, and RPD values of 0.92 g kg⁻¹, 0.40 g kg⁻¹, and 2.83, respectively. The research findings confirm the capabilities of FOD technology combined with the optimal band combination algorithm in optimizing spectral data, highlighting the advantages of multi-source data fusion modeling.

土壤有机质（SOM）是土壤肥力的重要组成部分，在作物生长和土壤改良中起着关键作用。可见和近红外光谱（Vis-NIR）是快速准确估计SOM含量的理想方法。本研究利用分数阶导数（FOD）和最优波段组合算法(二维光谱指数（2DSI）和三维光谱指数（3DSI）协同优化土壤可见光-近红外光谱数据，并结合环境变量（EV）建立SOM含量估算模型。结果表明，FOD的使用显著增强了光谱与SOM之间的相关性，特别是在0.4 ~ 1.4阶范围内。建模结果表明，与单独使用FOD构建的模型相比，将FOD与2DSI或3DSI相结合可以获得更好的结果。具体来说，与单独使用FOD相比，引入2DSI和3DSI分别使Rv2值增加了3.66%和8.54%。此外，当优化后的光谱数据与EV作为输入变量相结合时，模型的估计性能得到了显著提高。本研究采用2d + 3Ds + ev组合构建最优模型，Rv2、RMSEv和RPD值分别为0.92 g kg - 1、0.40 g kg - 1和2.83。研究结果证实了FOD技术结合最优波段组合算法在优化光谱数据方面的能力，凸显了多源数据融合建模的优势。

{"title":"Improved estimation of soil organic matter content in the Yellow River delta using spectral data optimization and environmental variables integration","authors":"Shugang Xie , Xueyuan Bai , Cheng Li , Xicun Zhu","doi":"10.1016/j.infrared.2026.106409","DOIUrl":"10.1016/j.infrared.2026.106409","url":null,"abstract":"<div><div>Soil Organic Matter (SOM) is a critical component of soil fertility, playing a key role in crop growth and soil improvement. Visible and near-infrared (Vis-NIR) spectroscopy is an ideal method for rapid and accurate estimation of SOM content. This study utilizes fractional-order derivative (FOD) and optimal band combination algorithm (two-dimensional spectral index (2DSI) and three-dimensional spectral index (3DSI)) collaboratively to optimize soil Vis-NIR spectral data and combines environmental variables (EV) to establish an estimation model for SOM content. The results indicate that the use of FOD significantly enhances the correlation between spectra and SOM, especially within the range of 0.4 to 1.4 orders. Modeling outcomes demonstrate that combining FOD with either 2DSI or 3DSI yields superior results compared to models constructed using FOD alone. Specifically, the introduction of 2DSI and 3DSI increases R<sub>v</sub>2 values by 3.66 % and 8.54 %, respectively, compared to using FOD alone. Furthermore, when optimized spectral data are combined with EV as input variables, the model’s estimation performance is markedly improved. The optimal model in this study is constructed using a combination of 2Ds + 3Ds + EVs, achieving R<sub>v</sub><sup>2</sup>, RMSE<sub>v</sub>, and RPD values of 0.92 g kg<sup>−1</sup>, 0.40 g kg<sup>−1</sup>, and 2.83, respectively. The research findings confirm the capabilities of FOD technology combined with the optimal band combination algorithm in optimizing spectral data, highlighting the advantages of multi-source data fusion modeling.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"154 ","pages":"Article 106409"},"PeriodicalIF":3.4,"publicationDate":"2026-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146074067","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}

引用次数: 0

High - throughput thermal NDT of pharmaceutical plastic container packaging 医药塑料容器包装的高通量热无损检测

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-17 DOI: 10.1016/j.infrared.2026.106407

A.O. Chulkov , V.P. Vavilov , O.A. Makushev , V.I. Stasevskii , A.V. Koldomov , A.N. Avstrievskikh

A novel application of active thermal nondestructive testing (TNDT) for the rapid detection of packaging defects in containers used for dietary supplements is described. The combination of convective and low-power optical heating enabled the detection of the following defects: delaminations (air bubbles) and loose shrinkage of the thermal film, crumpling, delaminations and contamination of the label, container contamination and cross-threaded cap sealing. It has been established that specific defect types produce characteristic thermal anomalies, allowing for the classification of detected thermal indications according to defect type. The TNDT productivity may reach 3–4 containers per second.

主动热无损检测（TNDT）的一种新的应用快速检测包装缺陷的容器用于膳食补充剂描述。对流和低功率光学加热的结合可以检测到以下缺陷：热敏膜的分层（气泡）和松散收缩，标签的皱缩，分层和污染，容器污染和交叉螺纹盖密封。已经确定，特定的缺陷类型产生特征热异常，允许根据缺陷类型对检测到的热指示进行分类。TNDT生产率可达3-4个容器/秒。

引用次数: 0

Infrared hyperspectral imaging integrated with an attentive spatial-spectral neural network for precise postharvest detection of Aspergillus flavus contamination and nutrient variations in peanut kernels 红外高光谱成像与空间光谱神经网络相结合，用于花生籽粒中黄曲霉污染和营养变化的采后精确检测

IF 3.4 3区物理与天体物理 Q2 INSTRUMENTS & INSTRUMENTATION

Infrared Physics & Technology

Pub Date : 2026-01-15 DOI: 10.1016/j.infrared.2026.106402

Zhen Guo , Yifei Qin , Xijun Shao , Fernando A. Auat-Cheein , Lianming Xia , Yemin Guo , Xia Sun , Fangling Du

Peanut kernels are highly susceptible to Aspergillus flavus contamination, posing significant food safety risks due to aflatoxin B₁ accumulation. This study applied infrared hyperspectral imaging, specifically covering visible-near infrared (VNIR, 400–1000 nm) and short-wave infrared (SWIR, 1000–2500 nm), to investigate the micro-interaction mechanisms between Aspergillus flavus and peanut kernels, focusing on spatio-temporal nutrient consumption and toxin accumulation. Infrared-based generalized two-dimensional correlation spectroscopy revealed a phased nutrient utilization strategy employed by Aspergillus flavus, identifying critical contamination phases at day 3 and day 5. An innovative attentive spatial-spectral synergy network (AS3Net) integrated with a novel bi-dimensional focus ripple module (BFRM), and significantly enhanced the prediction accuracy of moisture, protein, and oil contents in peanut kernels, achieving coefficient of determination of validation values of 0.932, 0.859, and 0.786, respectively. Ablation experiments highlighted that the combined use of spatial discovery, spectral insight modules, and dual fusion strategies improved model robustness, especially in predicting moisture content. Additionally, the AS3Net-BFRM framework provided a rapid, accurate classification of fungal-contaminated kernels with 100% accuracy. This advanced infrared hyperspectral imaging and deep learning approach presents a scalable, non-destructive, and efficient solution for real-time fungal contamination detection, which is crucial for enhancing food safety and managing aflatoxin risks in agricultural products.

花生仁极易受到黄曲霉污染，由于黄曲霉毒素B1的积累，对食品安全构成重大风险。本研究利用红外高光谱成像技术，特别是覆盖400-1000 nm的可见-近红外（VNIR）和1000-2500 nm的短波红外（SWIR），研究黄曲霉与花生籽粒的微观相互作用机制，重点研究营养物质消耗和毒素积累的时空变化。基于红外的广义二维相关光谱揭示了黄曲霉采用的分阶段养分利用策略，确定了第3天和第5天的关键污染阶段。创新的关注空间-光谱协同网络（AS3Net）与新型二维聚焦波纹模块（BFRM）相结合，显著提高了花生籽粒水分、蛋白质和油脂含量的预测精度，确定系数分别为0.932、0.859和0.786。消融实验表明，空间发现、光谱洞察模块和双融合策略的结合使用提高了模型的鲁棒性，尤其是在预测水分含量方面。此外，AS3Net-BFRM框架提供了快速、准确的真菌污染的分类，准确率为100%。这种先进的红外高光谱成像和深度学习方法为实时真菌污染检测提供了一种可扩展、非破坏性和高效的解决方案，这对于提高食品安全和管理农产品中的黄曲霉毒素风险至关重要。

{"title":"Infrared hyperspectral imaging integrated with an attentive spatial-spectral neural network for precise postharvest detection of Aspergillus flavus contamination and nutrient variations in peanut kernels","authors":"Zhen Guo , Yifei Qin , Xijun Shao , Fernando A. Auat-Cheein , Lianming Xia , Yemin Guo , Xia Sun , Fangling Du","doi":"10.1016/j.infrared.2026.106402","DOIUrl":"10.1016/j.infrared.2026.106402","url":null,"abstract":"<div><div>Peanut kernels are highly susceptible to <em>Aspergillus flavus</em> contamination, posing significant food safety risks due to aflatoxin B<sub>1</sub> accumulation. This study applied infrared hyperspectral imaging, specifically covering visible-near infrared (VNIR, 400–1000 nm) and short-wave infrared (SWIR, 1000–2500 nm), to investigate the micro-interaction mechanisms between <em>Aspergillus flavus</em> and peanut kernels, focusing on spatio-temporal nutrient consumption and toxin accumulation. Infrared-based generalized two-dimensional correlation spectroscopy revealed a phased nutrient utilization strategy employed by <em>Aspergillus flavus</em>, identifying critical contamination phases at day 3 and day 5. An innovative attentive spatial-spectral synergy network (AS3Net) integrated with a novel bi-dimensional focus ripple module (BFRM), and significantly enhanced the prediction accuracy of moisture, protein, and oil contents in peanut kernels, achieving coefficient of determination of validation values of 0.932, 0.859, and 0.786, respectively. Ablation experiments highlighted that the combined use of spatial discovery, spectral insight modules, and dual fusion strategies improved model robustness, especially in predicting moisture content. Additionally, the AS3Net-BFRM framework provided a rapid, accurate classification of fungal-contaminated kernels with 100% accuracy. This advanced infrared hyperspectral imaging and deep learning approach presents a scalable, non-destructive, and efficient solution for real-time fungal contamination detection, which is crucial for enhancing food safety and managing aflatoxin risks in agricultural products.</div></div>","PeriodicalId":13549,"journal":{"name":"Infrared Physics & Technology","volume":"154 ","pages":"Article 106402"},"PeriodicalIF":3.4,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023728","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}

引用次数: 0