Full-color 3D neural holography via stochastic physically consistent light field
Jiaqi Li, Yanan Zhang, Hebin Chang, Tianyou Li, Xing Fu, and Xingpeng Yan
Recent advancements in computer-generated holography have demonstrated that integrating neural networks can significantly enhance the speed and quality of multidepth hologram generation for complex 3D scenes. The inherent ill-conditioned nature of the mapping between diffraction fields and holograms poses a challenge, which neural networks adeptly address by nonlinearly approximating diffraction field patterns to rapidly generate holograms that meet stringent constraints. In particular, the lack of physical interpretability in phase-to-hologram mappings, the scarcity of high-quality 3D datasets, and the inefficiency of current learning strategies collectively hinder the reconstruction quality and broader applicability of neural holography. Herein, we present a stochastic physically consistent light field to address the aforementioned limitations. As revealed by mutual information analysis, our stochastic method decouples intensity-depth correlations. Benefiting from the stochastically generated spectrum-tunable intensity information, uniformly distributed depth information, and physically consistent phase, the neural network can predict ultra-multidepth, extended-depth-of-field, and full-color holograms trained on limited-depth, narrow-depth-of-field, and single-wavelength data without pre-existing training datasets. The resulting neural network model, comprising merely 712 parameters, achieves 4K full-color holographic encoding with 256 depth layers at a frame rate of 52.9 frames per second, yielding an average peak signal-to-noise ratio of 34.91 dB.Recent advancements in computer-generated holography have demonstrated that integrating neural networks can significantly enhance the speed and quality of multidepth hologram generation for complex 3D scenes. The inherent ill-conditioned nature of the mapping between diffraction fields and holograms poses a challenge, which neural networks adeptly address by nonlinearly approximating diffraction field patterns to rapidly generate holograms that meet stringent constraints. In particular, the lack of physical interpretability in phase-to-hologram mappings, the scarcity of high-quality 3D datasets, and the inefficiency of current learning strategies collectively hinder the reconstruction quality and broader applicability of neural holography. Herein, we present a stochastic physically consistent light field to address the aforementioned limitations. As revealed by mutual information analysis, our stochastic method decouples intensity-depth correlations. Benefiting from the stochastically generated spectrum-tunable intensity information, uniformly distributed depth information, and physically consistent phase, the neural network can predict ultra-multidepth, extended-depth-of-field, and full-color holograms trained on limited-depth, narrow-depth-of-field, and single-wavelength data without pre-existing training datasets. The resulting neural network model, comprising merely 712 parameters, achieves 4K full-color holographic encoding with 256 depth layers at a frame rate of 52.9 frames per second, yielding an average peak signal-to-noise ratio of 34.91 dB.
- May. 11, 2026
- Advanced Photonics
- Vol. 8, Issue 3, 036008 (2026)
- DOI：10.1117/1.AP.8.3.036008
Bifunctional metasurface for polarization transformation and vortex generation
Ziyin Xu, Min Liu, Xianhui Zhang, and Zhengyong Song
Metasurface, as an innovative 2D material, is promoting the technological development of optical control through its extraordinary capacity to precisely manipulate electromagnetic wavefront. We introduce a bilayer metasurface utilizing vanadium dioxide (VO2), which has the characteristic of a reversible insulator-to-metal transition. This metasurface utilizes the mutual interference of a diatomic structure to induce phase delay, enabling the transformation of an arbitrary polarized wave into a specified polarization. Leveraging propagation and geometric phases, it is designed to generate anomalously reflected vortex beams. Furthermore, by controlling the temperature, this approach enables selective switching of incident polarization between linear and circular states, as well as independent control over vortex beams. Specifically, when the temperature exceeds 68°C, the electromagnetic wave is modulated by VO2 patches. Utilizing the propagation phase, the y-polarized reflection port produces a vortex beam carrying topological charge (TC) l = 1, exhibiting an anomalous reflection at 30 deg. As the temperature is below 68°C, the gold patch array modulates the incident wave. Leveraging the geometric phase, the metasurface generates an anomalous vortex beam possessing TC l = - 2 and anomalous reflection at -30 deg in the right-handed circularly polarized reflection channel. This innovative metasurface opens up possibilities for reconfigurable optical devices and dynamic vortex control, holding great promise for applications in optical communication and information processing.Metasurface, as an innovative 2D material, is promoting the technological development of optical control through its extraordinary capacity to precisely manipulate electromagnetic wavefront. We introduce a bilayer metasurface utilizing vanadium dioxide (VO2), which has the characteristic of a reversible insulator-to-metal transition. This metasurface utilizes the mutual interference of a diatomic structure to induce phase delay, enabling the transformation of an arbitrary polarized wave into a specified polarization. Leveraging propagation and geometric phases, it is designed to generate anomalously reflected vortex beams. Furthermore, by controlling the temperature, this approach enables selective switching of incident polarization between linear and circular states, as well as independent control over vortex beams. Specifically, when the temperature exceeds 68°C, the electromagnetic wave is modulated by VO2 patches. Utilizing the propagation phase, the y-polarized reflection port produces a vortex beam carrying topological charge (TC) l = 1, exhibiting an anomalous reflection at 30 deg. As the temperature is below 68°C, the gold patch array modulates the incident wave. Leveraging the geometric phase, the metasurface generates an anomalous vortex beam possessing TC l = - 2 and anomalous reflection at -30 deg in the right-handed circularly polarized reflection channel. This innovative metasurface opens up possibilities for reconfigurable optical devices and dynamic vortex control, holding great promise for applications in optical communication and information processing.
- May. 11, 2026
- Advanced Photonics Nexus
- Vol. 5, Issue 4, 046006 (2026)
- DOI：10.1117/1.APN.5.4.046006
Unveiling the potential of ultra-thin perovskite-silicon tandem photovoltaics boosted with photonic management
Mónica Dyreby, Eva Almeida, Rui Gonçalves, Miguel Alexandre, Ivan M. Santos, Elvira Fortunato, Rodrigo Martins, Hugo Águas, and Manuel J. Mendes
The merging of thin-film photovoltaic (PV) technologies with tandem architectures, such as perovskite-on-silicon double-junction solar cells, offers avenues to expand solar electricity. Their combination of flexibility, affordability, low weight, and high efficiency enables applications ranging from portable electronics to building-integrated and vehicle-integrated PV, or solar-powered space systems. Nevertheless, the efficiency of perovskite-on-silicon tandem PV is often constrained by suboptimal optical management, particularly in ultra-thin designs where light trapping (LT) and current/voltage matching are critical. Here, we develop an optoelectronic framework to optimize 2- and 4-terminal perovskite-silicon tandem cells, featuring 1 μm-thick crystalline silicon absorbers with front-integrated photonic structures. To maximize power conversion efficiency (PCE), both the LT geometry and perovskite thickness were systematically optimized. In addition, indium tin oxide (ITO)-based and optically engineered interlayers are shown to exhibit similar optical performance, reinforcing ITO as a choice for 2-terminal tandems. The ultra-thin photonic-enhanced 2-terminal tandem achieves a PCE of 23.8%, corresponding to a 21.8% relative improvement over its planar counterpart, whereas the 4-terminal configuration reached a combined efficiency of 26.7%, primarily from LT-improved silicon photocurrent. These findings highlight the role of opto-electronically optimized light-management solutions in unlocking the potential of ultra-thin tandem solar cells for flexible, high-efficiency, and energy harvesting, paving the way for next-generation photovoltaics.The merging of thin-film photovoltaic (PV) technologies with tandem architectures, such as perovskite-on-silicon double-junction solar cells, offers avenues to expand solar electricity. Their combination of flexibility, affordability, low weight, and high efficiency enables applications ranging from portable electronics to building-integrated and vehicle-integrated PV, or solar-powered space systems. Nevertheless, the efficiency of perovskite-on-silicon tandem PV is often constrained by suboptimal optical management, particularly in ultra-thin designs where light trapping (LT) and current/voltage matching are critical. Here, we develop an optoelectronic framework to optimize 2- and 4-terminal perovskite-silicon tandem cells, featuring 1 μm-thick crystalline silicon absorbers with front-integrated photonic structures. To maximize power conversion efficiency (PCE), both the LT geometry and perovskite thickness were systematically optimized. In addition, indium tin oxide (ITO)-based and optically engineered interlayers are shown to exhibit similar optical performance, reinforcing ITO as a choice for 2-terminal tandems. The ultra-thin photonic-enhanced 2-terminal tandem achieves a PCE of 23.8%, corresponding to a 21.8% relative improvement over its planar counterpart, whereas the 4-terminal configuration reached a combined efficiency of 26.7%, primarily from LT-improved silicon photocurrent. These findings highlight the role of opto-electronically optimized light-management solutions in unlocking the potential of ultra-thin tandem solar cells for flexible, high-efficiency, and energy harvesting, paving the way for next-generation photovoltaics.
- May. 11, 2026
- Advanced Photonics Nexus
- Vol. 5, Issue 3, 036016 (2026)
- DOI：10.1117/1.APN.5.3.036016
Power Doppler-guided cross-modal fusion for augmented-view photoacoustic tomography
Zheng Qu, Xuanhao Zhang, Bin Ouyang, Cong Mai, Xu Tang, and Lidai Wang
Photoacoustic computed tomography (PACT) integrates optical excitation with ultrasound detection to deliver high-resolution imaging in deep tissues. A persistent limitation is the restricted angular coverage in linear array clinical systems, which results in image loss of vascular detail. Deep learning approaches have been investigated, yet models trained only on PACT data often fail to reconstruct vessels that are poorly represented due to incomplete angular sampling and low signal-to-noise ratio. A Doppler-enhanced photoacoustic network (DEPANet) is introduced that fuses co-registered ultrasound power Doppler with multi-wavelength PACT to generate enhanced structural and functional images. Ultrasound power Doppler, less sensitive to the limited-view problem, is acquired interleaved and spatially co-registered with PACT without additional hardware modifications. Validation is conducted in simulations, phantoms, small animals, and a human feasibility study. In phantom experiments, DEPANet expands the effective angular coverage more than 10-fold. In rats, DEPANet reconstructs obliquely oriented vessels, increases contrast-to-noise ratio by over 7 dB, and stabilizes hemoglobin oxygenation estimates. In human anterior interosseous artery and superior thyroid vasculature imaging, DEPANet recovers vessel morphology and provides physiologically consistent blood oxygen saturation maps. These findings demonstrate that DEPANet enables high-quality vascular and functional imaging using standard linear-array PACT systems.Photoacoustic computed tomography (PACT) integrates optical excitation with ultrasound detection to deliver high-resolution imaging in deep tissues. A persistent limitation is the restricted angular coverage in linear array clinical systems, which results in image loss of vascular detail. Deep learning approaches have been investigated, yet models trained only on PACT data often fail to reconstruct vessels that are poorly represented due to incomplete angular sampling and low signal-to-noise ratio. A Doppler-enhanced photoacoustic network (DEPANet) is introduced that fuses co-registered ultrasound power Doppler with multi-wavelength PACT to generate enhanced structural and functional images. Ultrasound power Doppler, less sensitive to the limited-view problem, is acquired interleaved and spatially co-registered with PACT without additional hardware modifications. Validation is conducted in simulations, phantoms, small animals, and a human feasibility study. In phantom experiments, DEPANet expands the effective angular coverage more than 10-fold. In rats, DEPANet reconstructs obliquely oriented vessels, increases contrast-to-noise ratio by over 7 dB, and stabilizes hemoglobin oxygenation estimates. In human anterior interosseous artery and superior thyroid vasculature imaging, DEPANet recovers vessel morphology and provides physiologically consistent blood oxygen saturation maps. These findings demonstrate that DEPANet enables high-quality vascular and functional imaging using standard linear-array PACT systems.
- May. 11, 2026
- Advanced Photonics Nexus
- Vol. 5, Issue 3, 036015 (2026)
- DOI：10.1117/1.APN.5.3.036015
Probing short-lived nuclear isomers in laser-induced plasmas
Youjing Wang, Weifu Yin, Yi Yang, Yixin Li, Kai Zhao, Zhiguo Ma, Guo-Qiang Zhang, Changbo Fu, and Yu-Gang Ma
With the advancement of high-intensity laser (HIL) technology, laser-induced plasma can produce short-lived nuclear isomers, which hold significant research value in fields such as nuclear-excitation mechanisms, nuclear clocks and radioactive medicine. However, due to intense electromagnetic pulses (EMPs) and X-rays, the detection of the short-lived isomers is still challenging today. To address this, an optical-fiber-coupled scintillator detection method is proposed in this study. The method can overcome the dilemma that traditional real-time detection methods face when struggling with the complex electromagnetic and radiation environment generated by HIL experiments, enabling real-time detection of characteristic signals on the nanosecond time scale during experiments. Employing a PW-level femtosecond laser-pumping ${}^{83}$ Kr to the $7/2+$ metastable state, which has a half-life of 156.94 ns, the de-excitation gamma-rays were detected successfully by the proposed detection system for the first time. This method addresses critical challenges in EMP-dominated HIL environments, enables investigations of ultra-fast nuclear processes and further advances experiments related to high-repetition-rate intense lasers.With the advancement of high-intensity laser (HIL) technology, laser-induced plasma can produce short-lived nuclear isomers, which hold significant research value in fields such as nuclear-excitation mechanisms, nuclear clocks and radioactive medicine. However, due to intense electromagnetic pulses (EMPs) and X-rays, the detection of the short-lived isomers is still challenging today. To address this, an optical-fiber-coupled scintillator detection method is proposed in this study. The method can overcome the dilemma that traditional real-time detection methods face when struggling with the complex electromagnetic and radiation environment generated by HIL experiments, enabling real-time detection of characteristic signals on the nanosecond time scale during experiments. Employing a PW-level femtosecond laser-pumping ${}^{83}$ Kr to the $7/2+$ metastable state, which has a half-life of 156.94 ns, the de-excitation gamma-rays were detected successfully by the proposed detection system for the first time. This method addresses critical challenges in EMP-dominated HIL environments, enables investigations of ultra-fast nuclear processes and further advances experiments related to high-repetition-rate intense lasers.
- May. 11, 2026
- High Power Laser Science and Engineering
- Vol. 14, Issue 2, 02000e23 (2026)
- DOI：10.1017/hpl.2026.10111
Sulfur-vacancy-enabled broadband near-infrared emission from wurtzite CdS quantum dots synthesized via a high-concentration precursor route
Jinke Bai, Haodi Duan, Zhelei Ren, Xiaomin Xie, Peng Wang, Xilong Liang, Junli Wang, Jiarui Hao, Bing Xu, Shuwen Xue, and Xiao Jin
Harnessing defect emission in CdS quantum dots (QDs) for efficient near-infrared (NIR) emission remains difficult. We report a synergistic strategy combining crystal-phase and defect engineering. High-concentration synthesis stabilizes the metastable wurtzite phase. A sulfur-rich stoichiometry (Cd:S=1:1.25) increases the concentration of sulfur vacancies (V_S), which is electron paramagnetic resonance (EPR)-confirmed and enables broadband NIR emission at 740 nm. Trace Zn2+ incorporation, confirmed as surface-adsorbed, further enhances monodispersity and passivates non-radiative traps, boosting emission intensity. A prototype light-emitting diode (LED) integrated with the Zn2+-modified CdS QDs exhibits broad emission (580–1100 nm), demonstrating dual functionality for visible lighting and NIR night vision.Harnessing defect emission in CdS quantum dots (QDs) for efficient near-infrared (NIR) emission remains difficult. We report a synergistic strategy combining crystal-phase and defect engineering. High-concentration synthesis stabilizes the metastable wurtzite phase. A sulfur-rich stoichiometry (Cd:S=1:1.25) increases the concentration of sulfur vacancies (V_S), which is electron paramagnetic resonance (EPR)-confirmed and enables broadband NIR emission at 740 nm. Trace Zn2+ incorporation, confirmed as surface-adsorbed, further enhances monodispersity and passivates non-radiative traps, boosting emission intensity. A prototype light-emitting diode (LED) integrated with the Zn2+-modified CdS QDs exhibits broad emission (580–1100 nm), demonstrating dual functionality for visible lighting and NIR night vision.
- May. 11, 2026
- Chinese Optics Letters
- Vol. 24, Issue 5, 051601 (2026)
- DOI：10.3788/COL202624.051601
Development and validation of an airborne high-spectral-resolution lidar system for ocean optical property profiling
Hui Qi, Yan He, Weibiao Chen, Chunhe Hou, Jian Ma, Sheng Su, Peng Chen, Guangli Yu, Fu Yang, Xiaoquan Song, Qi Chen, Huixin He, and Xinke Hao
Vertical profiling of seawater optical properties is important for understanding marine subsurface processes. We developed an airborne high-spectral-resolution lidar (HSRL) system that combined a frequency-stabilized narrow-linewidth laser, iodine absorption filtering, and hybrid analog–photon-counting detection to achieve wide-dynamic-range ocean observations. A flight experiment near Sanya, South China Sea, was conducted with synchronous shipborne measurements for validation. The lidar-retrieved diffuse attenuation coefficient Kd and particulate backscattering coefficient bbp agree well with in situ results, with mean absolute percentage deviations of 4.23% and 13.98% and corresponding root mean square errors of 0.0030 and 1.63 × 10-4 m-1, respectively. The results verify the capability of the developed airborne HSRL system for accurate ocean optical property profiling.Vertical profiling of seawater optical properties is important for understanding marine subsurface processes. We developed an airborne high-spectral-resolution lidar (HSRL) system that combined a frequency-stabilized narrow-linewidth laser, iodine absorption filtering, and hybrid analog–photon-counting detection to achieve wide-dynamic-range ocean observations. A flight experiment near Sanya, South China Sea, was conducted with synchronous shipborne measurements for validation. The lidar-retrieved diffuse attenuation coefficient Kd and particulate backscattering coefficient bbp agree well with in situ results, with mean absolute percentage deviations of 4.23% and 13.98% and corresponding root mean square errors of 0.0030 and 1.63 × 10-4 m-1, respectively. The results verify the capability of the developed airborne HSRL system for accurate ocean optical property profiling.
- May. 11, 2026
- Chinese Optics Letters
- Vol. 24, Issue 5, 050101 (2026)
- DOI：10.3788/COL202624.050101
Single-shot scattering imaging beyond the optical memory effect via speckle-sparsity decoupling
Xiyuan Luo, Juncheng Liu, Xin Wang, Jiawei Li, Meng Xiang, Dong Wang, Jingbo Duan, Tong Zhang, Xue Dong, Zihan Geng, and Fei Liu
Overcoming the optical memory effect range to achieve large field-of-view imaging through scattering media without prior information has remained a significant challenge. We present a single-shot large field-of-view imaging technique based on the spatial sparsity characteristics theory of speckle pattern, enabling blind reconstruction of multiple targets beyond the optical memory effect range without prior information or wavefront modulation. The theoretical innovation lies in revealing the spatially sparse distribution of speckles when multiple isolated targets exceed the optical memory effect separation distance. By establishing a sparse mapping relationship between scattering transmission and object space, the method achieves unsupervised decoupling of low-cross-talk speckle regions while simultaneously reconstructing target intensity and positional information. Combined with the modified phase retrieval algorithm, the complete scene of multiple targets beyond the optical memory effect range is reconstructed. Experiments show that under varying scattering media and spectral bandwidths, this approach achieves a field-of-view expansion exceeding 6.82 times that of conventional methods, with a relative localization accuracy of 97.5%. We present the first introduction of spatial sparsity concepts into speckle field analysis; it establishes a new theoretical framework for deep-tissue biological observation and optical sensing in complex environments.Overcoming the optical memory effect range to achieve large field-of-view imaging through scattering media without prior information has remained a significant challenge. We present a single-shot large field-of-view imaging technique based on the spatial sparsity characteristics theory of speckle pattern, enabling blind reconstruction of multiple targets beyond the optical memory effect range without prior information or wavefront modulation. The theoretical innovation lies in revealing the spatially sparse distribution of speckles when multiple isolated targets exceed the optical memory effect separation distance. By establishing a sparse mapping relationship between scattering transmission and object space, the method achieves unsupervised decoupling of low-cross-talk speckle regions while simultaneously reconstructing target intensity and positional information. Combined with the modified phase retrieval algorithm, the complete scene of multiple targets beyond the optical memory effect range is reconstructed. Experiments show that under varying scattering media and spectral bandwidths, this approach achieves a field-of-view expansion exceeding 6.82 times that of conventional methods, with a relative localization accuracy of 97.5%. We present the first introduction of spatial sparsity concepts into speckle field analysis; it establishes a new theoretical framework for deep-tissue biological observation and optical sensing in complex environments.
- May. 11, 2026
- Advanced Photonics
- Vol. 8, Issue 3, 036009 (2026)
- DOI：10.1117/1.AP.8.3.036009
Dual-wavelength dual combs driven by harmonic and fundamental mode-locking compound for absolute ranging
Jianing Tao, Churan Zhang, Qianqian Huang, Chengbo Mou, Minglie Hu, and Youjian Song
For dual-comb ranging, maintaining the large nonambiguity range under high sampling speed is challenging, hence hindering the ultimate performance of dual-comb sources. Here, by unlocking the dimension of repetition rate, we have demonstrated single-cavity dual combs consisting of fundamental mode-locking and harmonic mode-locking (HML), with an exemplified repetition rate of 50.8 and 305 MHz, respectively. The dual combs exhibit comparable noise performance to conventional twin laser combs, i.e., characterized by the inherent asynchronous sampling magnification of hybrid pulses, the temporal drifts of HML are less than 50 ps within a 500 ms timing window, and the relative period jitter of the dual combs is 1.42 fs. Using the dual combs, we achieved absolute distance measurement under an update time of 0.1 ms at a nonambiguity range of 2.95 m. In particular, the ranging precision reaches 745 nm at 200 ms acquisition time. Our findings represent the first dual-comb ranging implementation using the HML, achieving enhanced sampling speed without compromising the nonambiguity range. The results show that complex multipulse dynamics can be manipulated for high-speed and high-precision metrological applications. Nevertheless, our findings may enhance the performance of other dual-comb-enabled applications such as pump-probe and spectroscopy.For dual-comb ranging, maintaining the large nonambiguity range under high sampling speed is challenging, hence hindering the ultimate performance of dual-comb sources. Here, by unlocking the dimension of repetition rate, we have demonstrated single-cavity dual combs consisting of fundamental mode-locking and harmonic mode-locking (HML), with an exemplified repetition rate of 50.8 and 305 MHz, respectively. The dual combs exhibit comparable noise performance to conventional twin laser combs, i.e., characterized by the inherent asynchronous sampling magnification of hybrid pulses, the temporal drifts of HML are less than 50 ps within a 500 ms timing window, and the relative period jitter of the dual combs is 1.42 fs. Using the dual combs, we achieved absolute distance measurement under an update time of 0.1 ms at a nonambiguity range of 2.95 m. In particular, the ranging precision reaches 745 nm at 200 ms acquisition time. Our findings represent the first dual-comb ranging implementation using the HML, achieving enhanced sampling speed without compromising the nonambiguity range. The results show that complex multipulse dynamics can be manipulated for high-speed and high-precision metrological applications. Nevertheless, our findings may enhance the performance of other dual-comb-enabled applications such as pump-probe and spectroscopy.
- May. 11, 2026
- Advanced Photonics Nexus
- Vol. 5, Issue 4, 046005 (2026)
- DOI：10.1117/1.APN.5.4.046005
Non-destructive evaluation of the nanosecond laser damage characteristics of the Nd,Y:CaF2 crystal via photothermal weak absorption using a fully connected neural network
Chong Shan, Huamin Kou, Dapeng Jiang, Qinghui Wu, Zhonghan Zhang, Yanyan Xue, Lizhi Fang, Zhen Zhang, Rongrong Liu, Yafei Lian, Xing Peng, Yuanan Zhao, and Liangbi Su
Nd,Y:CaF2 (NYCF) crystals are exceptional gain materials for high-power laser drivers; however, laser-induced damage remains a substantial challenge that restricts their broader application. In this study, by establishing an in situ testing system for photothermal weak absorption and the laser-induced damage threshold (LIDT), the relationship between the photothermal weak absorption characteristics of NYCF and its LIDTs was analyzed. A fully connected neural network was employed to facilitate deep learning of these relationships, thereby enabling non-destructive evaluation of NYCF via photothermal weak absorption. Moreover, this study examined both the effect of spot size during testing and the influence of crystal orientation on the evaluation outcomes. The underlying mechanisms were further elucidated by investigating NYCF’s thermal mechanical properties and damage characteristics. This work not only offers a rapid, non-destructive method for evaluating the laser damage resistance of NYCF using artificial intelligence but also enhances the understanding of its damage mechanisms.Nd,Y:CaF2 (NYCF) crystals are exceptional gain materials for high-power laser drivers; however, laser-induced damage remains a substantial challenge that restricts their broader application. In this study, by establishing an in situ testing system for photothermal weak absorption and the laser-induced damage threshold (LIDT), the relationship between the photothermal weak absorption characteristics of NYCF and its LIDTs was analyzed. A fully connected neural network was employed to facilitate deep learning of these relationships, thereby enabling non-destructive evaluation of NYCF via photothermal weak absorption. Moreover, this study examined both the effect of spot size during testing and the influence of crystal orientation on the evaluation outcomes. The underlying mechanisms were further elucidated by investigating NYCF’s thermal mechanical properties and damage characteristics. This work not only offers a rapid, non-destructive method for evaluating the laser damage resistance of NYCF using artificial intelligence but also enhances the understanding of its damage mechanisms.
- May. 11, 2026
- High Power Laser Science and Engineering
- Vol. 14, Issue 2, 02000e24 (2026)
- DOI：10.1017/hpl.2026.10112
OpticsNatural ScienceEngineeringAgricultureMedical ScienceInterdisciplinaryMultidisciplinaryAll Subjects
Journal
Independent polarization and phase control in liquid crystal programmable THz metasurfaces
May. 03, 2026
Journal
Tunable emission directionality in transparent quantum-dot LEDs via photonic interface engineering
May. 03, 2026
Journal
Deep 3D laser micromachining enabled by kilohertz-rate adaptive energy-flux compensation
May. 03, 2026
Journal
May. 03, 2026
Special Issue on Debris produced in High Power Laser Interactions (2026)
Submission Open:23 April 2026; Submission Deadline: 31 August 2026
Special Issue on the 40th anniversary of National Laboratory on High Power Laser and Physics (2026)
Submission Open:15 April 2026; Submission Deadline: 15 August 2026
Special Issue on Forensic Document Examination (2024)
Published
Editor (s): Carolyne Lindsay Bird,Xu Yang
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20