最新刊期
- 摘要：Colorless polyimide （CPI） is a critical flexible substrate material in the fabrication of transparent flexible organic light-emitting diode（OLED） displays. To identify the cause of yellowing in fluorinated CPI on glass substrates and to improve its optical properties， this work systematically examines the mechanism and explores effective countermeasures. First， we comprehensively analyzed the fluorinated CPI samples using UV-Vis transmittance spectroscopy， scanning electron microscopy（SEM）， and time-of-flight secondary ion mass spectrometry（ToF-SIMS）. Meanwhile， inorganic barrier layers such as ITO， SiOx， and SiNx were deposited on glass surfaces by sputtering or chemical vapor deposition to evaluate their suppression effect on the yellowing phenomenon. The results reveal that the yellowing of fluorinated CPI arises from the migration of alkaline earth metal ions from the glass into the CPI layer， where they combine with fluorine to form charge-transfer complexes（CTCs）. This complexation significantly reduces the fluorine content in CPI， thereby promoting the yellowing of the material. In addition， the degree of yellowing differs between the air side and the glass side of the CPI film. With the introduction of inorganic barrier layers， the yellowing of fluorinated CPI was effectively suppressed， and fluorinated 9-μm-thick CPI films with a cutoff wavelength of 403 nm were successfully obtained. The migration of alkaline earth metal ions from the glass substrate is the primary cause of yellowing in fluorinated CPI， and the deposition of inorganic barrier layers can significantly improve its optical transparency， offering a viable pathway for the preparation of high-performance transparent flexible OLED substrate materials.关键词：transparent flexible display;organic light-emitting diode display;colorless polyimide;yellowing0|0|0更新时间：2026-11-09
- 摘要：High-resolution remote sensing images are one of the core data sources for supporting remote sensing applications such as land-cover monitoring， ecological assessment， disaster identification， and refined interpretation of ground objects. However， owing to the limitations of sensor spatial resolution， revisit period， imaging cost， and weather conditions， high-resolution images are sometimes difficult to obtain or cannot continuously cover the target study areas. To address this problem， this paper proposes FSSENet， a cross-sensor super-resolution reconstruction network based on frequency-spatial synergistic enhancement（FSSE）. The proposed network aims to recover high-quality high-resolution remote sensing images from low-resolution observations by jointly enhancing frequency-domain information and spatial-domain structural representation. FSSENet mainly consists of three innovative modules： 1） A frequency-spatial synergistic high- and low-frequency separation module is constructed. In this module， Fourier transform is first employed to explicitly separate high- and low-frequency components in the frequency domain. Meanwhile， convolution kernels with different receptive fields are introduced in the spatial domain to extract local high-frequency information and global low-frequency information. Then， self-attention is used to independently perform synergistic enhancement of high- and low-frequency features between the spatial and frequency domains， thereby obtaining higher-quality high-frequency and low-frequency representations and improving the model’s joint representation capability for main structures， edge contours， and texture details. 2） A low-frequency-gated enhanced fusion module is designed. This module first uses low-frequency features to generate channel and spatial gating weights， which enables adaptive selection and response enhancement of high-frequency features. The selected high-frequency details are then fused with low-frequency structural features through residual connections. In this way， the effective fusion of high- and low-frequency features is achieved， while noise interference is suppressed and edge and texture details are further strengthened. 3） A multi-scale pooling module with progressively enlarged receptive fields is proposed. By gradually expanding the receptive field， this module enhances the contextual modeling ability of the network for multi-scale ground objects and improves the preservation of spatial structures in complex remote sensing scenes. To verify the effectiveness of FSSENet， two cross-sensor datasets， RSRData-FHJ and RSRData-GH， are constructed for experiments， providing paired data support for model training and evaluation. Experimental results show that， compared with seven benchmark methods， FSSENet improves PSNR by 0.41 to 5.29 dB and SSIM by 0.115 0 to 0.176 7 on the RSRData-FHJ dataset. In addition， LPIPS and ERGAS also show significant improvements， demonstrating that the proposed method provides an effective solution for cross-sensor remote sensing image super-resolution reconstruction and practical remote sensing image enhancement tasks.关键词：cross-sensor remote sensing image;super-resolution;frequency-domain features;feature interaction0|0|0更新时间：2026-11-09
- 摘要：Small objects in remote sensing images are characterized by their limited scale， dense distribution， and complex backgrounds， posing significant challenges to object detection. To address the problems of feature attenuation， insufficient contextual and directional information modeling， and limited detection-head adaptability in single-stage detection frameworks for remote sensing small-object scenarios， a Low-Parameter Collaborative Directional Attention Network（LCDA-Net） is proposed. Built upon YOLOv11， the proposed method employs a low-parameter collaborative bidirectional path network to enhance the semantic and detail collaboration of multi-scale features， thereby mitigating the loss of small-object information during downsampling. A context-aware receptive field module is introduced to improve the structural and edge representations of remote sensing small objects by incorporating directional contextual modeling， dynamic receptive fields， channel recalibration， and detail preservation. In addition， a small-object-aware dynamic detection head is developed to achieve adaptive feature enhancement and task decoupling across hierarchical， spatial， and task dimensions. Experiments are conducted on publicly available small-object datasets， including VisDrone2019 and AI-TODv2. Experimental results show that， compared with YOLOv11-S， LCDA-Net improves the mAP50 from 39.3% to 48.1% on the VisDrone2019 dataset， representing an increase of 8.8 percentage points， while reducing the number of parameters by 47.8%. On the AI-TODv2 dataset， LCDA-Net achieves AP50 and AP50：95 of 64.0% and 29.4%， respectively， with the detection accuracy for very tiny （APvt）， tiny （APt）， and small （APs） objects improved by 8.1， 9.4， and 8.0 percentage points， respectively. These results demonstrate that LCDA-Net effectively enhances the detection of objects at different scales while achieving a favorable balance between detection accuracy and model complexity.关键词：small-object detection;remote sensing images;low-parameter detection network;multi-scale feature fusion;dynamic detection head0|0|0更新时间：2026-11-09
- 摘要：The electro-optical characteristics， response time， and viewing-angle characteristics of liquid crystal devices are typically measured using separate instruments. Repeated adjustments of the optical path and sample position are also required during measurement， making the process relatively complicated and increasing the difficulty of maintaining consistent test conditions across different measurements. To address these issues， an integrated test platform was designed and constructed for the performance characterization of liquid crystal devices. The platform integrates driving signal generation， transmitted light intensity acquisition， and automatic adjustment of sample orientation， allowing the above parameters to be measured using the same optical path. The platform was further applied to the performance testing and optimization of polymer-dispersed liquid crystal（PDLC） devices. The effects of polymer monomer combinations， the mass ratio of trimethylolpropane trimethacrylate （TMPTMA） to 2-ethylhexyl acrylate （2-EHA）， cell gap， and ultraviolet curing time on device performance were systematically investigated. After multiple rounds of screening， sample A6 had a contrast ratio of 2.33 in the initial monomer-combination screening stage. At a TMPTMA-to-2-EHA mass ratio of 1∶1， a cell gap of 10 μm， and an ultraviolet curing time of 15 min， the final sample D4 had a contrast ratio of 14.5， with a rise time of 2.9 ms and a fall time of 47.6 ms. Finally， viewing-angle characterization of the optimized device showed that its contrast-ratio distribution was relatively uniform across different viewing directions. In conclusion， the platform enables integrated measurement of multiple performance parameters of liquid crystal devices， avoids sample transfer and optical-path adjustment between different measurements， and can be used for the performance comparison and process optimization of liquid crystal devices. It provides a highly integrated and low-cost experimental solution for comprehensive performance testing of liquid crystal devices.关键词：liquid crystal device;integrated testing;polymer-dispersed liquid crystal;electro-optical characteristics;response time;viewing-angle characteristics1|0|0更新时间：2026-11-09
- 摘要：To address the high missed-detection rates and insufficient accuracy in UAV-based maritime small object detection caused by small scales and complex backgrounds， we propose SEW-YOLO， an improved YOLO11n-based algorithm. First， a Split Channel Attention Bottleneck（SCAB） module is designed in the backbone， enhancing small object feature extraction through channel grouping and coordinate attention mechanisms. Second， an Enhanced Spatial Pyramid Pooling Fast（SPPF_E） module replaces large-kernel pooling with small-kernel pooling for small object scenarios， using efficient channel attention to adaptively calibrate multi-scale receptive fields. Meanwhile， WIoU loss replaces the original CIoU to improve bounding box regression accuracy for small objects. Finally， ablation and comparative experiments on the publicly available SeaDronesSee dataset show that SEW-YOLO achieves mAP50 of 68.1% and mAP50-95 of 40.6%—improvements of 2.8 and 1.5 percentage points over the baseline， respectively—while maintaining model lightweight nature， validating the algorithm’s effectiveness.关键词：small object detection;YOLO11n;lightweight;maritime search and rescue;UAV imagery21|8|0更新时间：2026-11-04
- 摘要：To address the problems that insulator defects such as breakage and pollution flashover in UAV aerial inspection images are characterized by small target scale， weak edge textures， and strong background interference， which cause existing detection models to be prone to missed and false detections and make it difficult to balance detection accuracy with real-time performance， this paper proposes a lightweight insulator defect detection algorithm integrating multi-scale edge enhancement and quality awareness， named MEAQ-YOLOv11. A multi-scale edge difference enhancement module is designed in the backbone network， which strengthens the local edge and texture responses of defect regions through multi-scale adaptive pooling and high-frequency edge difference. In the neck network， an adaptive multi-branch scale fusion network is constructed， which enhances information interaction among features at different levels through multi-branch adaptive weighting and multi-scale depthwise convolution. Finally， a quality-aware lightweight shared detection head is designed， which uses shared convolution to reduce redundant parameters in the detection head while leveraging bounding box distribution information to estimate localization quality and refine classification scores. Experimental results on two public datasets， CPMID and IDID， show that compared with the baseline YOLOv11n， the proposed algorithm achieves a clear improvement in detection accuracy on both datasets， with mAP50 increasing by 4.54 percentage points on CPMID and by 1.29 percentage points on IDID. Meanwhile， the number of parameters is reduced to 2.58 M， the computational cost is 6.3 GFLOPs， the model size is 5.3 MB， and the frame rate increases from 235.0 FPS to 254.1 FPS， achieving model lightweighting and inference acceleration alongside the improved detection accuracy. While maintaining a lightweight structure and real-time performance， the proposed MEAQ-YOLOv11 effectively improves the detection accuracy and localization reliability of small insulator defects under complex backgrounds， providing a reference for intelligent defect identification in power inspection scenarios.关键词：YOLOv11;multi-scale edge enhancement;adaptive feature fusion;object detection;lightweight network30|9|0更新时间：2026-11-04
- 摘要：Target detection algorithms have been widely applied to defect detection in unmanned aerial vehicle（UAV） inspection images of transmission lines. However， existing methods remain limited in representing irregular defects during feature extraction. Large variations in target scales can cause fine details of small defects to attenuate during cross-layer feature fusion， while complex background interference can lead to confusion between defect features and background responses. To address these issues， a lightweight detection algorithm， termed CSF-DEIM， is proposed based on geometry awareness and a spatial-frequency gather-distribute mechanism. A lightweight geometry-aware module is constructed to redesign the backbone network， where dynamic geometric sampling and low-cost linear feature expansion are combined to improve the adaptive representation of irregular defects. A spatial-frequency collaborative gather-distribute network is then developed based on the gather-distribute mechanism， incorporating wavelet transform and multi-scale progressive channel attention to strengthen cross-scale feature interaction and reduce the loss of fine details of small defects. Furthermore， an efficient discriminative frequency-domain feedforward network is introduced into the intra-scale feature interaction module， where spatial-domain local enhancement and frequency-domain adaptive filtering are employed to improve defect discrimination under complex backgrounds. Experimental results show that CSF-DEIM achieves an mAP@0.5 of 92.6%， representing an improvement of 4.3 percentage points over the baseline model， while reducing the number of parameters to 3.37 M. Supplementary experiments on a public dataset further verify the applicability and generalization capability of the proposed model under different data distributions. On the NVIDIA Jetson Xavier NX platform， CSF-DEIM achieves a real-time inference speed of 42.3 FPS， enabling lightweight， high-accuracy， and real-time defect detection for transmission-line inspection.关键词：transmission line inspection;defect detection;DEIM;geometry awareness;spatial-frequency collaboration;gather-distribute mechanism27|7|0更新时间：2026-11-04
- 摘要：Electrochromic technology shows broad application prospects in smart windows， displays， and anti-glare rearview mirrors owing to its low driving voltage and low power consumption. Liquid crystals （LCs）， combining the fluidity of liquids with the anisotropy of crystals， provide a unique platform for multidimensional optical modulation through their electric-field responsiveness and optical anisotropy. Recently， the integration of LCs and electrochromism is reshaping the research landscape of smart light modulation. This review focuses on the material systems， device architectures， and multifunctional integration of LC-based electrochromic smart windows and display devices. First， the basic structures and performance parameters of electrochromic devices are introduced. Then， the discussion is organized according to the two roles that LCs play in the devices： （i） as functional enhancement components， where LCs do not participate in electrochromic reactions but provide light-scattering modulation， structural color， or polarization control， enabling dual-functional integration of electrically controlled light modulation and electrochromic coloration， as well as color gamut expansion and information encryption through cholesteric LCs and cellulose nanocrystals； and （ii） as electrochromic active components， where LC themselves participate in redox reactions or form charge-transfer complexes with electrolyte ions to produce color changes， including LC/ammonium salt systems and LC/ionic liquid systems. Finally， the current mechanistic controversies， performance bottlenecks， and future development directions are discussed， aiming to provide a reference for the commercialization of high-performance， low-cost， multifunctional LC-based electrochromic devices.关键词：liquid crystals;electrochromism;multifunctional devices28|29|0更新时间：2026-11-02
- 摘要：Cholesteric liquid crystals（CLCs）， characterized by periodic helical superstructures， exhibit selective light reflection and dynamically tunable photonic bandgaps， making them promising platforms for smart optical materials and tunable photonic devices. By incorporating photoresponsive molecular switches， the molecular organization and helical superstructures of CLCs can be regulated at multiple levels through mechanisms including photoinduced configurational changes， chirality transfer， molecular motion， and photothermal effects. This review systematically summarizes the molecular design strategies， photoresponsive mechanisms， and structural regulation principles of azobenzene-， spiropyran-， diarylethene-， hydrazone-， and overcrowded-alkene molecular motor-based systems， as well as other emerging photoresponsive systems. Particular emphasis is placed on their applications in smart displays， chiral luminescence， information encryption， adaptive optics， and soft actuation. The advantages and limitations of different photoresponsive systems are comparatively analyzed in terms of response wavelength， tuning range， cycling stability， and device integration. Finally， future perspectives are discussed with a focus on visible-light responsiveness， energy utilization efficiency， multimodal responses， and scalable device fabrication， providing a reference for the further development of photoresponsive CLCs materials.关键词：photoresponsive molecules;cholesteric liquid crystals;structural color;molecular switches;helical twisting power49|20|0更新时间：2026-10-26
- 摘要：In response to the challenges associated with the physical coupling of color and brightness in traditional planar additive mixed-color electrowetting display electronic paper， including significant light leakage in the dark state and the limitations of post-processing algorithms imposed by the device's bottom layer， we propose a dynamic filtering type color electrowetting display electronic paper device along with a two-stage brightness maximum compensation color conversion driving algorithm. This innovative device substitutes the fixed filter with controllable color electro-ink wet ink and incorporates independently driven black electro-ink wet units for each RGB sub-pixel， thereby achieving physical decoupling of color and brightness within each channel.To facilitate accurate color image reproduction， we designed a two-stage color conversion algorithm that prioritizes the maximization of brightness， leveraging the unique characteristics of this dynamic filtering architecture. In the first stage， the black unit is fully activated to optimize the target color of the color voltage. The second stage maintains the color voltage while optimizing the voltage of the black unit to achieve brightness compensation， introducing a channel penalty term to mitigate color deviation.We constructed a forward optical virtual simulation platform based on the measured spectra of each monochromatic pixel. The results of the simulation comparison indicate that， in contrast to single-layer color and traditional image processing algorithms， this collaborative device-algorithm scheme effectively suppresses color distortion， enhances the ability to reproduce details in dark areas， and achieves a balance between color accuracy and output brightness. This research offers both theoretical and simulation insights for the driving design of high-quality color electrowetting display electronic paper.关键词：electrowetting display;electronic paper;dynamic color filter;color‑luminance decoupling28|18|0更新时间：2026-10-26
- 摘要：The low contrast ratio of LCoS panel is a critical factor limiting their large-scale commercial deployment. Achieving high contrast has become a pressing challenge for LCoS devices in industrial applications.This study proposes a contrast enhancement method employing an electrically tunable liquid crystal retarder. A dynamically tunable liquid crystal retarder was introduced into the optical path preceding the LCoS panel. By simultaneously modulating the amplitude and frequency of the applied voltage， dynamic compensation for the residual retardation of the liquid crystal within the LCoS panel is realized. Furthermore， the dependence of the contrast ratio on the control voltage parameters is systematically investigated and experimentally validated. An intrinsic contrast ratio measurement system was constructed using a commercial light source optical engine and an LCoS panel. The testing system was subsequently improved to enable contrast measurement with a retarder. The experimental results demonstrate that the contrast ratio of the LCoS panel can be significantly enhanced by modulating the amplitude and frequency of the voltage applied to the liquid crystal retarder. The optimal contrast ratio achieved was 340.5， representing a 9.12 fold improvement over the intrinsic contrast ratio of 37.32. Frequency modulation significantly refuces the driving voltage of the liquid crystal retarder， thereby lowering the overall power consumption of the LCoS optical engine. This study provides a theoretical basis and technical support for the multi-degree-of-freedom modulation of electrically tunable liquid crystal retarders in LCoS optical engine applications.关键词：liquid crystal optics;LCoS panel;electronically controlled liquid crystal retarder;dynamic tuning;contrast ratio;drive frequency33|23|0更新时间：2026-10-26
- 摘要：To address the limitations of existing Transformer-based image super-resolution methods in modeling local textures and high-frequency details， an Efficient Mixed Transformer with Wavelet （W-EMT） is proposed by integrating wavelet transform into the Transformer architecture. The discrete Haar wavelet transform is introduced into the Transformer framework to perform multi-level frequency decomposition of input features， enabling multi-scale frequency-domain representation learning. In local feature modeling， a local wavelet mixer is constructed to decompose features into different frequency subbands， where low-frequency structural information and high-frequency texture information are enhanced using a differentiated strategy， followed by feature reconstruction via the inverse wavelet transform. In global feature modeling， a stripe-window self-attention mechanism is introduced to capture long-range dependencies along the horizontal and vertical directions. The proposed mechanism maintains global contextual modeling capability while reducing computational complexity and enables collaborative learning of local and global features. The model is trained on the DIV2K and Flickr2K datasets and evaluated on the Set5， Set14， BSD100， Urban100， and Manga109 benchmark datasets using peak signal-to-noise ratio （PSNR） and structural similarity index （SSIM） as evaluation metrics. Experimental results demonstrate that W-EMT achieves consistent performance improvements across all test datasets. For ×4 image super-resolution， W-EMT improves PSNR by 0.06 dB and SSIM by 0.0011 on the Urban100 dataset compared with EMT， demonstrating superior high-frequency detail reconstruction in complex texture regions and verifying the effectiveness of the proposed method.关键词：image super-resolution;wavelet transform;Transformer;Multi-scale features;attention mechanism21|17|0更新时间：2026-10-26
- 摘要：Electrowetting displays（EWDs）suffer from oil film splitting， while the optimization of driving waveforms still relies heavily on empirical parameter tuning. To address these issues， a closed-loop driving waveform optimization method based on image feedback and improved Bayesian optimization is proposed. A physical-state-aware mechanism is introduced to dynamically switch acquisition functions， while a worst-value penalty for oil film splitting and a neighborhood exclusion strategy are incorporated to treat failure states as optimization constraints， thereby enabling automatic search of driving parameters. Experimental results show that the optimized waveform achieves a pixel aperture ratio of 57.38% without obvious oil film splitting， representing improvements of 10.36 and 10.71 percentage points over the conventional PWM rectangular waveform and gradient-rising waveform， respectively. The average luminance reaches approximately 730 cd/m² with improved output stability. Ablation experiments further demonstrate the contributions of different driving stages to reducing transient electric-field impact， stabilizing the oil boundary， and improving oil recovery. Cross-device experiments indicate that directly transferring an optimized waveform to other devices results in performance degradation， whereas the proposed method can re-optimize the driving parameters according to device-specific response characteristics. These results demonstrate that the proposed method can automatically obtain high-performance and safe feasible driving parameters within a limited experimental budget， thereby improving the automation， efficiency， and reliability of EWD driving waveform optimization.关键词：electrowetting display;oil film splitting;closed-loop optimization33|27|0更新时间：2026-10-26
- 摘要：Reconfigurable multifunctional integration is an important research direction of terahertz devices. However， the functions of existing devices are often solidified， and it is difficult to achieve flexible regulation of different functions in the same structure. We design a liquid-crystal-tunable THz 1 bit coding metasurface capable of reconfigurable applications for beam steering and near-field pattern imaging. The metasurface adopts a metal-liquid crystal-metal reflection unit structure， the reflection phase and amplitude are controlled by the characteristics of the liquid crystal dielectric constant changing with the bias voltage. At the array coding level， the cross-electrode row-column coding method is adopted. The coding state of the array unit is determined by the modular two-plus operation of the upper and lower electrode coding matrices， and the dynamic reconstruction of the array coding distribution is realized. The simulation results show that a beam deflection of up to 46.5 ° can be achieved by changing the period of the coding sequence at 0.680 THz. When only one layer of the metal electrode is set to the coding sequence and the other layer is set to the fixed coding state， the switching from the x-direction dual beam to the y-direction dual beam can be realized. When the coding sequence is set at the upper and lower metal electrode layers at the same time， the reconfigurable control from two beams to four beams can be realized. In addition， the near-field reconstruction of “T，” “H，” and “Z” letter patterns at z=855 μm is realized by using the difference of reflection amplitude of the unit structure under different bias conditions at 0.524 THz. The designed liquid crystal tunable terahertz coding metasurface can realize the integration of beam control and pattern imaging in the same structure， which provides an implementation scheme for terahertz wireless communication， near-field imaging and multi-functional reconfigurable device design.关键词：terahertz;liquid crystal;coding metasurface;beam steering;near-field imaging22|16|0更新时间：2026-10-26
-
Performance investigation of novel indium tin oxide targets for high-precision display applications AI导读
摘要：This study systematically compares the material properties， thin-film performance， and application performance in display devices between the new water-less indium tin oxide （W-ITO） target and normal indium tin oxide （N-ITO） target. Material analysis reveals that W-ITO mitigates particle contamination induced by water during annealing by introducing trace amounts of metallic elements at 0.10 at.%. Additionally， its sputter-resistant secondary phase area and particle size （2.02 μm， 16.80%） are significantly smaller than those of N-ITO， effectively reducing arc generation from the source. In terms of performance after magnetron sputtering， W-ITO exhibits superior average sheet resistance， excellent resistivity stability， high crystalline quality， and high transmittance. It also demonstrates faster wet etching rates， cleaner surfaces， and reduced undercutting， facilitating the fabrication of high-precision patterns. In thin-film transistor （TFT） device applications， devices based on W-ITO electrodes exhibit standard electrical characteristics， with smooth and distortion-free transfer curves indirectly confirming the high-quality interface between the electrode and active layer. After further integration and encapsulation into glass panels， the final product demonstrates outstanding performance in optical contrast， gamma curve， image retention， electrostatic discharge tolerance， and power consumption， validating the enhancement of comprehensive performance and reliability of display panels by W-ITO electrodes. Finally， mass production data confirms that using W-ITO targets significantly improves the first-pass yield of ITO deposition batches （~0.14%）， reduces the rework rate for secondary deposition defects （~0.37%）， and effectively extends equipment maintenance cycles， delivering substantial economic benefits. This study provides systematic experimental evidence and theoretical support for material selection and process optimization of high-reliability， high-performance ITO.关键词：indium tin oxide;magnetron sputtering;thin film transistor;glass panel;yield55|61|0更新时间：2026-10-16 - 摘要：Liquid crystal photonic crystals are a class of soft-matter photonic materials based on self-assembled helical structures of liquid crystal molecules， distinguished by the selective Bragg reflection and geometric phase modulation arising from their chiral helical architectures. Unlike conventional rigid photonic crystals， the periodicity of liquid crystal photonic crystals originates from the rotational order of molecular directors rather than the translational order of molecular centers. This fundamental distinction endows them with two core advantages： large-area self-assembly fabrication without the need for complex nanolithography， and external-field tunability enabled by weak intermolecular interactions. This review systematically surveys the evolutionary trajectory of liquid crystal photonic crystals from structural color displays to planar optics. First， external-field tuning strategies for structural color displays are summarized from three perspectives—wavelength， amplitude， and chirality. Subsequently， the physical origins and implementation schemes of geometric phase in cholesteric liquid crystals and blue phase liquid crystals are elaborated， revealing the intrinsic logic underlying the transition from spectral modulation to wavefront engineering. On this basis， recent advances in frontier research are highlighted， including structured light generation and manipulation， dynamic holographic displays， augmented reality waveguides， optical computing， and quantum optics， illustrating the structure-phase-function relationships. Finally， key challenges and potential solutions toward the practical deployment of liquid-crystal photonic crystal planar optics are briefly discussed from the perspectives of material performance optimization and device structural innovation. It is anticipated that liquid crystal photonic crystals， as a planar optical platform combining tunability with multidimensional light-field manipulation capabilities， will play a significant role in next-generation photonics and quantum optical devices.关键词：cholesteric liquid crystals;blue phase liquid crystals;structural color;geometric phase;planar optics37|51|0更新时间：2026-10-16
-
Research on the cascaded sequence of liquid crystal polarization grating in wide-angle scanning AI导读
摘要：In wide-angle beam-steering applications employing cascaded liquid crystal polarization gratings （LCPGs）， diffraction efficiency exhibits significant angular dependence， leading to substantial fluctuations in system-level energy stability. To address this critical limitation， this study systematically investigates the role of the cascading sequence in modulating the diffraction efficiency uniformity of a two-dimensional beam-deflection architecture， with the objective of establishing an optimal design paradigm for achieving highly uniform diffracted energy distribution. An optimization framework is developed based on a uniformity evaluation factor as the objective function， and the performance of devices under both ideal and realistic liquid crystal waveplate conditions is compared to elucidate the underlying mechanisms governing the influence of cascade order on efficiency uniformity. Our results reveal that， in a cascaded system employing liquid crystal waveplates with superior modulation performance， placing a large-angle grating at the front end of the cascaded chain can effectively suppress the edge attenuation of the diffraction efficiency distribution. For a field of regard of 50°×50°， the proposed sequence optimization reduces the maximum total diffraction efficiency discrepancy between the central and peripheral regions from 40.86% to 13.71% in transmit mode， and from 44.05% to 15.36% in receive mode. This work not only provides clear theoretical guidelines for designing liquid crystal polarization grating systems with high energy uniformity over wide steering angles， but also demonstrates the compatibility of the proposed optimization strategy with shared-aperture transmit-receive architectures. These findings hold significant promise for advancing compact beam-steering technologies in applications such as laser communication and lidar.关键词：liquid crystal polarization grating;beam steering;cascade sequence;diffraction efficiency;wide-angle scanning44|44|0更新时间：2026-10-16 - 摘要：To address the issues of abnormal liquid crystal alignment and efficiency loss induced by direct immersion of subwavelength gold gratings in liquid crystal， we propose a novel anchoring scheme using polymethyl methacrylate （PMMA） as the photoresist for gold grating patterning， the trench-filling material， and the alignment layer for the upper substrate of the liquid crystal cell. This design simplifies the fabrication process， realizes uniform liquid crystal alignment， and avoids the liquid crystal infiltration problem. To reveal the anchoring behavior of subwavelength gold gratings in direct contact with liquid crystal and clarify the performance limitations of the direct infiltration scheme， we first fabricated two groups of liquid crystal cells and experimentally investigated the molecular alignment via cross-polarized optical microscopy. Two groups of liquid crystal cells are fabricated， with the alignment direction of the bottom substrate under the gold grating was set parallel and perpendicular to the grating lines， respectively. Note that liquid crystal molecules align along the grating lines on the subwavelength gold grating samples， which induces a twisted molecular configuration in the samples. Such an alignment is not the preferred parallel anchoring mode for classic phase modulator technologies. Furthermore， its orientation is orthogonal to the polarization direction of surface plasmon polaritons（SPPs） excited by the gold grating resonance， so the grating resonance cannot be effectively tuned by the variation of liquid crystal refractive index to achieve dynamic tuning. Based on the interfacial alignment characteristics observed in the above experiments， TechWiz 3D software is further used to simulate the changes in the liquid crystal director on the lower surface of the gold grating under power conditions for two schemes： filling the gold grating groove with PMMA and liquid crystal.， and the corresponding effective refractive index variation of liquid crystal at the grating lower surface is obtained. The derived effective refractive index data are then imported into FDTD Solutions software to compare the reflection phase modulation and reflectivity characteristics between two schemes： gold grating trenches filled with PMMA and filled with liquid crystal. Simulations demonstrate that compared with the liquid crystal filling scheme， the PMMA filling scheme achieves a more complete
- 摘要：Conventional switchable liquid crystal （LC） gratings suffer from polarization sensitivity， poor continuous electrical tunability of grating period， and slow response， which cannot meet the requirements for multi-dimensional and high precision light manipulation in complex optical systems. To address these drawbacks， this work adopts a flexoelectric LC mixture with a broad operating temperature range （0~65 ℃）. Combined with patterned photoalignment， we fabricate a dual-mode switchable LC grating capable of continuous tuning of grating period， diffraction angle， and diffraction efficiency. The grating contains two periodic structures with markedly different feature sizes and can generate two distinct diffraction signals at different propagation distances. At zero applied voltage， transverse one-dimensional diffraction with small diffraction angles originating from the photo aligned pattern is observed. When the applied voltage exceeds 14 V， cross-stripe diffraction of large angles originating from the flexoelectric effect is activated. Within the voltage range of 14 V to 26 V， the grating period can be continuously tuned from 6.65 µm to 1.87 µm. The tuning for full range of period takes approximately 100 s， and the tuning speed is nearly 12 times higher than that of conventional light-driven cholesteric LC gratings. Meanwhile， the device exhibits weak polarization dependence and supports simultaneous continuous electrical modulation of grating period， diffraction angle， and diffraction efficiency. It features a simple fabrication procedure without elaborate patterned electrode structures. It holds promising application prospects in dynamic optical scenarios such as spectral scanning， adaptive beam shaping， and augmented reality displays， and provides a new technical route for the design of tunable LC optoelectronic devices.关键词：flexoelectric effect;broad temperature range;continuous period tuning;mode-switching grating;patterned photoalignment32|13|0更新时间：2026-09-21
- 摘要：To address the issue of declining local evidence quality in Internet of Vehicles collaborative perception， which affects the stability and usability of collaborative fusion results， the fusion process is modeled as local perception evidence reliability estimation， and a spatiotemporal consistency-based robust fusion method （STC-RF） is proposed for collaborative perception. First， low-dimensional semantic evidence is constructed using a target existence heatmap. Subsequently， by comprehensively considering cross-vehicle spatial consistency， historical occupancy continuity， and negative evidence conflict， BEV grid-level local evidence reliability is generated and injected as a bias into the intermediate fusion of attention to continuously adjust the fusion contributions of different collaborators at various spatial positions. Furthermore， a conservative historical occupancy update is adopted to ensure that the historical time reference continuously participates in the fusion process， reducing the impact of local evidence fluctuations on the temporal continuity of the collaborative results. Experimental results demonstrate that under the condition of background evidence degradation， the proposed method improves AP@0.7 from 82.09% to 86.20%， reduces SSR from 8.21% to 5.53%， and decreases CFNR from 0.90% to 0.72%， indicating that the method enhances the fusion stability under non-ideal collaborative perception conditions.关键词：internet of vehicles;collaborative perception;perception evidence degradation;spatiotemporal consistency;robust fusion9|11|0更新时间：2026-09-21
- Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China 130033 Postal code：130033
- Tel：0431-86176059 Email：yjxs@ciomp.ac.cn
- Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0