最新刊期
卷 33 ， 期 17 ， 2025
- 摘要：Extreme ultraviolet lithography， a pivotal technology in semiconductor manufacturing， imposes atomic-scale precision requirements on optical surfaces and necessitates negligible surface and subsurface damage. This paper addresses ultra‑precision fabrication of optical components by first elucidating the mechanisms by which material properties and manufacturing processes induce surface errors. Technical challenges and recent advances in measuring surface errors across spatial scales are then analyzed. For low‑spatial‑frequency figure errors， emphasis is placed on absolute measurement methods and sub‑aperture stitching techniques. For mid-to high-spatial-frequency roughness， developments in white‑light interferometry microscopy， atomic force microscopy， and super‑resolution white‑light interferometry are reviewed. To meet multi‑modal characterization requirements for surface and subsurface defects， the advantages and limitations of complementary detection methods-including electron microscopy， interferometry， scanning probe techniques， and scattering approaches-are comparatively assessed. Finally， against the backdrop of prevailing technical bottlenecks and the demand for high‑performance optics， future directions are delineated， highlighting intelligent measurement systems， multi‑physical coupled characterization， and in‑situ monitoring. This analysis is intended to serve as a technical reference to support domestic production of critical equipment components.关键词：advanced optical manufacturing;optical measurement;surface metrology;surface roughness measurement;defect measurement1241|5438|1更新时间：2026-09-11
- 摘要：An adaptive spot-centroid extraction method is proposed to address low extraction accuracy and insufficient real-time performance in cooperative-target feature-point imaging for large-span dynamic measurement， problems that arise from pronounced size variation and edge jitter. Exploiting the motion coherence of cooperative targets， a dynamic ROI feature-parameter model is first established to enable rapid and accurate ROI localization via interframe motion prediction， thereby substantially reducing the data volume for subsequent processing. Canny edge-detection parameters are then adaptively adjusted using an Otsu-based threshold optimization strategy， which improves noise suppression across varying measurement distances while enhancing computational efficiency. Sub-pixel edge localization is refined by combining a multi-directional Sobel operator with Zernike moments enhanced by an intensity ramp， and centroid coordinates are obtained via a robust least-squares circle fitting method improved through Gauss-Newton iteration. Validation on simulated image datasets and experimental measurements demonstrates that， for small-scale spots with simulated edge blur， the centroid positioning error of the proposed method ranges from 0.001 to 0.025 pixels under different noise levels. In practical tests， the ROI prediction algorithm satisfies measurement scenarios with accelerations up to 8.75 m/s2， and the repeatability error of spot positioning at measurement distances of 10-30 m remains stable between 0.016 and 0.040 pixels， outperforming conventional methods. Meanwhile， spot-extraction speed is increased by approximately 75.5%， markedly improving real-time processing capability. The proposed approach offers effective technical support for cooperative-target measurement applications.关键词：cooperative target detection;adaptive spot centroid extraction;inter-frame motion prediction;improved Sobel-Zernike moment586|694|0更新时间：2026-09-11
-
Modeling and calibration of system errors in distance-angle fusion spatial coordinate measurement AI导读
摘要：Facing multi-objective， autonomous， parallel measurement demands in the manufacturing of high-end equipment （e.g.， aviation and aerospace） and the construction and operation of large scientific facilities （e.g.， large radio telescopes）， improvements are presented for the distance-angle fusion spatial coordinate measurement system previously proposed. The geometric-structure error model has been refined， and a high-precision control-field calibration method based on an adaptive weighted optimization algorithm is introduced. Precise calibration of the system's error parameters was achieved. In indoor calibration experiments， the average deviation of coordinates across 24 calibration points was 0.098 1 mm， closely matching the theoretical average deviation from single-measurement simulation. In outdoor astronomical observation conditions， the standard deviation of point-position measurements for 10 observation points was better than 0.021 mm， the maximum tangential displacement comparison deviation of the reflective surface was 0.022 mm， and the maximum gravitational-direction displacement deviation was 0.040 mm. These results demonstrate that the proposed method effectively calibrates the error parameters of the distance-angle fusion coordinate measurement system and satisfies the real-time control requirements for active surface shape accuracy in astronomical observations.关键词：coordinate measurement;Error modeling;error calibration;multi-objective444|470|0更新时间：2026-09-11 - 摘要：Zoom optical systems realize multiple focal lengths within a single configuration by actively and controllably adjusting optical elements， enabling flexible imaging modalities. Reflective and catadioptric zoom systems are particularly suited to high-resolution imaging applications owing to their capacity for large apertures and extended focal lengths. This paper presents a systematic review of the development of reflective and catadioptric zoom optics， delineates the evolutionary characteristics at each stage， and classifies the structural typologies of representative systems. Their technical attributes and performance metrics are analyzed and compared. Finally， prospects for future advances in these systems are discussed to stimulate interdisciplinary dialogue among researchers in optical design.关键词：zoom optical systems;reflective zoom;catadioptric zoom;optical system design570|754|1更新时间：2026-09-11
- 摘要：Accurate monitoring of plant chlorophyll content is essential for agricultural management and physiological research. Traditional methods typically require contact measurements， are time-consuming， and cannot readily capture three-dimensional structural information. To address these limitations， a fluorescence LiDAR system （SLiDAR） based on the Scheimpflug principle was developed for non-contact， high-precision measurement of chlorophyll fluorescence， enabling simultaneous acquisition of fluorescence spectra and three-dimensional （3D） structure. The SLiDAR employs a continuous-wave 450 nm laser and a tilted CMOS sensor architecture. Normalization of the red-channel fluorescence signal by the blue-channel elastic scattering signal was implemented to suppress environmental interference and enhance signal-to-noise ratio （SNR）. Validation against a handheld chlorophyll meter （SPAD-502PLUS） for various leaf samples demonstrated a strong linear correlation （R²>0.98）， confirming reliable chlorophyll fluorescence retrieval and millimeter-level spatial resolution at ranges of 5-10 m. The results indicate that the fluorescence SLiDAR offers high spectral sensitivity and spatial resolution， enabling efficient， non-destructive indirect monitoring of chlorophyll content via fluorescence measurements. This approach provides an innovative technical solution for assessing plant physiological status， with potential for enhanced applicability in agricultural remote sensing through multi-wavelength excitation and algorithmic optimization.关键词：Scheimpflug LiDAR;fluorescence imaging;3D point cloud;chlorophyll fluorescence360|573|0更新时间：2026-09-11
- 摘要：A long-wave infrared time-sharing polarization imaging system was developed to address instability in polarization response during high-speed imaging and to enable detection of UAV targets against strong infrared clutter. An optimal polarization step angle was identified through analysis and angle-error model simulations to minimize polarization error. An adaptive polarization-angle estimation algorithm， based on a time-varying parametric Kalman filter， was implemented to provide accurate angle estimates and maintain a stable polarization response at high frame rates. Experimental results at a 50 Hz imaging rate demonstrate that the filtering algorithm reduces polarization response fluctuations by 91%， achieves a polarization accuracy of 0.50%， and attains a polarization-angle accuracy of 1.52°. Additionally， the target-region contrast in polarization images is enhanced by 3-10×. The proposed imaging system significantly improves UAV target contrast and satisfies detection requirements in environments with strong infrared clutter.关键词：imaging systems;infrared polarization imaging;Kalman filter;contrast improvement;background suppression556|819|0更新时间：2026-09-11
Modern Applied Optics
- 摘要：To achieve high-quality polishing of zinc sulfide optical components， this study develops a theoretical and experimental framework for robotic ultrasonic-assisted polishing， with the aim of establishing a high-precision， high-efficiency polishing theory and process tailored to zinc sulfide. A contact model between abrasive grains and the workpiece during robotic ultrasonic-assisted polishing is first established， and the three-dimensional trajectory of abrasive grains under ultrasonic vibration is characterized through kinematic analysis. Based on Preston's material removal theory， the material removal function is modified by introducing a dynamic pressure term and a relative velocity term induced by ultrasonic vibration， thereby constructing a zinc sulfide material removal model that accounts for high-frequency vibration effects. Ultrasonic-assisted stagnation-point polishing experiments are performed to determine the effects of polishing pre-pressure and spindle speed on material removal depth and removal rate for zinc sulfide， and the modified removal function is validated against experimental data. Finally， using the modified removal function and optimized process parameters， comparative experiments between ultrasonic-assisted and conventional robotic polishing of zinc sulfide flats are conducted. Surface form accuracy and surface roughness before and after polishing are quantified by white-light interferometry and profilometry. Results indicate that robotic ultrasonic-assisted polishing yields improvements in both form and roughness relative to non-ultrasonic-assisted polishing： the surface roughness Sa decreases from 2.632 nm to 1.782 nm， and the form error PV decreases from 0.206λ to 0.149λ （λ=632.8 nm）. These findings demonstrate that robotic ultrasonic-assisted polishing， through the synergistic effects of ultrasonic vibration and optimized polishing trajectory， provides superior polishing performance for high-quality processing of zinc sulfide.关键词：robot polishing;ultrasonic-assisted polishing;zinc sulfide;remove function;surface quality390|618|0更新时间：2026-09-11
- 摘要：Conventional industrial CT in situ loading devices suffer from thermal instability， control hysteresis， and thermal gradient artifacts at extreme temperatures， impairing observation of thermomechanical coupling in composite materials. To address these limitations， a purpose built bidirectional temperature controlled in situ loading apparatus for industrial CT systems was developed and implemented. The apparatus enables non destructive characterization of the dynamic microstructural evolution of composite specimens within a controlled temperature range of -20 °C to 150 °C by optimizing the fixture geometry， incorporating a symmetric upper and lower bidirectional temperature control module， and integrating a highly adaptable in situ tensile control system. The device was evaluated through in situ tensile tests on carbon fiber reinforced polymer （CFRP） plate specimens across multiple temperatures. Coupled with industrial CT three dimensional reconstruction， the mesoscale damage evolution of CFRP was captured in real time. Results demonstrate that， throughout the -20 °C to 150 °C range， incremental tensile displacement can be continuously recorded by industrial CT from micro damage initiation through to macroscopic failure. At elevated temperatures， matrix softening， pronounced interface debonding， and thermal expansion mismatch were observed， resulting in reduced mechanical performance. At low temperatures， resin embrittlement， microcrack initiation， and interface stress concentration were evident， leading to performance degradation. These findings confirm the apparatus provides reliable temperature control and loading stability across the tested range. The developed device offers a robust technical platform for elucidating damage mechanisms and structural evolution of composite materials under extreme thermal conditions and establishes an experimental foundation for their engineering applications in aerospace， transportation， and related fields.关键词：Industrial CT;in-situ tensile loading device;bidirectional temperature control;carbon fiber reinforced polymer;internal damage evolution375|701|0更新时间：2026-09-11
- 摘要：To address the limited adaptability and single-functionality of traditional single-robot systems in complex amphibious tasks， a novel bionic amphibious mother-child robot system is proposed. The mother robot features a retractable， turtle-shell-shaped spherical enclosure and integrates four mechanical legs with water-jet propulsion units. The child robot comprises a microstreamlined cabin equipped with a propeller thruster and an underwater short-range visible-light communication module. A gear-meshing mount-and-release mechanism is developed to enable reliable deployment and retrieval of the child robot. To avoid Euler-angle singularities， a six-degree-of-freedom fully coupled dynamic model is formulated using unit quaternions， and a UQ-PID control algorithm is introduced to coordinate the eight servomotors and four water propulsion units of the mother robot. Fluid simulations indicate that the spherical structure effectively reduces hydrodynamic disturbance and enhances stability. Experimental results demonstrate that the mother robot achieves approximately 60 cm forward locomotion on land within 20 s and maintains a precise straight-line trajectory underwater with minimal deviation and stable thrust. The child robot exhibits a pool trajectory error within ±0.30 m and attains a peak speed of 32.51 cm/s. The presented system and methods constitute a feasible technical approach for coordinated amphibious operation of mother-child robotic platforms.关键词：bionic amphibious spherical mother robot;sub-robot;motion control;fluid simulation519|645|1更新时间：2026-09-11
Micro\/Nano Technology and Fine Mechanics
- 摘要：Faint astronomical target detection is fundamental to space resource exploration， development， and utilization， providing essential information for celestial object tracking， asteroid mining， and deep-space mission planning. Detecting such targets remains challenging due to minimal pixel coverage， low signal-to-noise ratios （SNR）， and sparse intrinsic features， which contribute to high false-negative rates and limited performance in existing approaches. To address these limitations， a Hierarchical Attention Dense Nested Network （HADN-Net） is proposed， a deep-learning framework specifically designed to enhance detection accuracy for faint targets in astronomical imagery.HADN-Net incorporates two novel modules to tackle the principal challenges of faint target detection. The Multi-branch Ghost-Dilated Attention Module （Mul-GDAM） adopts a multi-receptive-field feature extraction strategy that combines dilated convolutions with ghost modules to expand contextual modeling while preserving computational efficiency. This module effectively captures local neighborhood features， suppresses complex background noise， and strengthens the discriminative representation of faint targets. The Hierarchical Feature Aggregation Module （HiFAM） employs cross spatial-channel attention mechanisms to dynamically fuse multi-scale features. By adaptively weighting feature maps across resolutions， HiFAM preserves faint-target information and mitigates the scale-variation problem common in astronomical observations.Extensive experiments on real ground-based optical images demonstrate that HADN-Net achieves state-of-the-art performance， yielding a recall of 94.648%， precision of 95.518%， and an F1-score of 95.081%， and outperforming five leading methods. In particular， recall is improved by 2.855% relative to the second-best method， confirming the effectiveness of the proposed approach for faint celestial target detection. This work introduces a scalable solution applicable to space debris monitoring， exoplanet discovery， and near-Earth object （NEO） tracking.关键词：computer vision;ground-based optical astronomical images;faint astronomy;target detection;neural networks313|611|0更新时间：2026-09-11
- 摘要：Time-series photometric signals from spaceborne targets contain rich information about their operational states. Conventional short-time Fourier transform （STFT） uses a fixed window length， which limits time-frequency resolution and reduces effectiveness for target classification and identification. To address this limitation， an eigenfrequency inversion method based on an adaptive-window-length STFT is proposed. A hybrid loss function—combining minimum information entropy， peak energy contrast， and background noise level—is formulated to drive a dynamic window-length adjustment strategy， enabling adaptive matching between window parameters and signal nonstationarity. A dedicated simulation apparatus was developed to acquire time-series photometric signals that reflect space-target operational characteristics. By varying the momentum-wheel rotational states， scenarios of on-orbit steady operation and attitude adjustment were simulated， producing high-rate photometric signals induced by surface vibration. Experimental results show that the proposed algorithm achieves high frequency resolution， with eigenfrequency inversion precision of 0.4 Hz， and dynamically adjusts window length to provide multi-scale temporal resolution. The steady-state eigenfrequency of momentum wheels was estimated with error below 0.3 Hz， and transition points associated with changes in operational conditions were accurately identified. The method offers substantial value for assessing space-target operational state， monitoring health status， and providing early fault warnings for attitude-control systems.关键词：space targets;temporal luminosity;adaptive window length;short-time Fourier Transorm;inversion of frequency characteristics356|552|0更新时间：2026-09-11
- 摘要：The performance of vehicle inertial navigation systems degrades severely in GNSS-denied environments. To address this limitation， a starlight–inertial integrated navigation method based on satellite celestial observation is proposed. Targets （stars and satellites） within the sensor field of view are first identified through image preprocessing. Attitude is then determined from star measurements obtained by the star sensor. Using satellite ephemerides， the satellite position in the Earth-centered， Earth-fixed （ECEF） frame at the current epoch is computed， and the resulting geometric vector between vehicle and satellite is formed to establish the measurement equation. By combining this measurement with an inertial navigation error model， a state equation is constructed and position and attitude estimates are updated via Kalman filtering. Compared with conventional star-only observations， the relatively limited range to satellites permits direct provision of position information from satellite observations， preventing divergence of navigation position error and enhancing practical applicability. The effect of satellite position error is analyzed through simulation， which demonstrates that the proposed integrated navigation algorithm achieves high accuracy. The algorithm was further validated in a static navigation test， yielding a 2-hour navigation position error of 41.42 m； across tests with varying initial position errors， navigation position error remained below 150 m.关键词：satellite observation;starlight-inertial integrated navigation;kalman filtering;error analysis708|692|1更新时间：2026-09-11
Information Sciences
- Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China Postal code：130033
- Tel：0431-86176855 Email：gxjmgc@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