最新刊期
卷 33 ， 期 19 ， 2025
- 摘要：To improve target positioning accuracy in UAV-based electro-optical imaging and to inform localization methodology and system design， an electro-optical imaging target localization model is developed. Spatial geometric positioning is performed via coordinate transformation， and error sensitivity is quantified using error synthesis theory. Model accuracy is evaluated through Monte Carlo simulations combined with experimental measurements， with emphasis on the coupling effects of angular positioning errors. Monte Carlo results indicate that， as the number of measurements increases， positioning errors in the geodetic Cartesian coordinate system converge to a normal distribution with zero mean. The spatial positioning error is reduced from 18.71 m for a single measurement to 0.28 m with multiple measurements， approaching the actual measurement value. Analysis of accuracy-influencing factors shows that the effects of positioning and ranging errors are predominantly determined by their magnitudes， whereas the impacts of angular and attitude errors are also strongly dependent on the measurement state. This study therefore quantifies the coupling of multi-source heterogeneous errors， clarifies the mechanisms by which distinct error sources affect positioning accuracy， and provides a theoretical foundation for developing error suppression strategies in UAV electro-optical imaging localization， thereby contributing to improved target positioning and measurement precision.关键词：photoelectric imaging localization;positioning accuracy;target localization model;error analysis20|14|0更新时间：2025-10-31
- 摘要：To achieve rapid and precise birefringence microscopy， a photoelastic-modulation-based imaging scheme is proposed. Leveraging the high modulation frequency and purity of photoelastic modulators， birefringence imaging measurements are realized. The disparity between the modulator's high-frequency operation and the camera's low frame rate is resolved via stroboscopic illumination. The sample's birefringence phase retardation distribution is retrieved from only three images acquired at photoelastic modulation phases of 0°， 30°， and 90°. An experimental system was constructed and tested on a wave plate and linden stem sections. Results indicate a wave-plate retardation error below 1%， retardation accuracy within ±λ/300， and a maximum retardation-distribution deviation not exceeding 1.62 nm. Distinct retardation regions in the linden stem sections enabled differentiation of anatomical parts， with single-measurement acquisition time under 3 ms. The method achieves high-speed， high-precision mapping of birefringence phase retardation and offers an advanced technical approach for studies of two-dimensional materials， biological tissue imaging， and related fields.关键词：microscopy imaging;polarization modulation;stroboscopic illumination;phase retardation;birefringence13|5|0更新时间：2025-10-31
- 摘要：Traditional laser-induced breakdown spectroscopy （LIBS） employing low-repetition-rate solid-state lasers exhibits insufficient stability during extended continuous operation. Although fiber-laser-induced breakdown spectroscopy （FL-LIBS） offers superior long-term stability， its high repetition rate poses challenges for conventional quantitative approaches， which typically rely on time-integrated spectra and calibration with reference samples. To remove dependence on reference standards， a high-speed， high-repetition-frequency triggered-gating system was implemented on an FL-LIBS platform to capture transient plasma spectra. A calibration-free quantitative analysis algorithm was developed， and the Internal Reference Self-absorption Correction （IRSAC） method was improved to compensate for spectral self-absorption， thereby enhancing the accuracy of calibration-free quantification. Experimental results for TC4 titanium alloy indicate total analysis distance errors （DE） ranging from 0.883% to 3.928%（mass fraction）， corresponding to reductions of 16.879% to 26.597 %（mass fraction） relative to uncorrected self-absorption. These findings demonstrate that high-accuracy calibration-free quantitative analysis can be achieved under high-repetition-frequency fiber laser ablation， offering a viable approach for LIBS systems that require prolonged continuous operation.关键词：laser induced breakdown spectroscopy;fiber laser;quantitative analysis;self-absorption effect;calibration-free method18|6|0更新时间：2025-10-31
- 摘要：In response to the need for high-precision geometric feature measurement under complex intervisibility conditions in large-scale advanced precision manufacturing， a hidden-point coordinate measurement method based on inertial-visual fused attitude estimation is proposed. The system architecture and measurement principles for hidden-point coordinate determination are analyzed， and Monte Carlo simulations are conducted to identify the principal factors influencing measurement uncertainty. An attitude-fusion approach employing spherical linear interpolation （slerp） for quaternions is introduced to optimize interpolation paths during quaternion fusion， addressing challenges associated with attitude interpolation. On this basis， an inertial–visual fused attitude estimation algorithm is developed. Calibration procedures for the parameters of the hidden-point measurement system are then formulated. An experimental setup is implemented to validate the effectiveness and reliability of the proposed method through parameter calibration and hidden-point coordinate measurement experiments. Experimental results indicate that the repeated calibration deviation of the coordinate transformation matrix remains below 0.03°， and the repeated calibration deviation of the structural parameter of the hidden-point target remains below 45 μm， confirming the stability and reliability of the calibration procedures. Based on the calibration results， the inertial–visual fusion method reduces the mean hidden-point coordinate measurement error by 60.63% relative to the monocular-vision method， with the maximum spatial coordinate error kept within 130 μm over a 10 m range. These findings demonstrate that the proposed method meets the geometric measurement requirements of typical high-precision manufacturing environments and exhibits substantial practical potential.关键词：monocular vision;hidden point coordinate measurement;inertial measurement unit;fused attitude estimation12|5|0更新时间：2025-10-31
Modern Applied Optics
- 摘要：To resolve the conflict between large-stroke motion and high-precision measurement， a novel cross-scale micro–nano coordinate measuring machine （CSMN-CMM） with a 100 mm × 100 mm × 100 mm measuring volume is presented. Unlike conventional coordinate measuring machines （CMMs）， the CSMN-CMM adopts an independent metrology system layout that satisfies the Abbe principle in all three dimensions， thereby isolating the metrology system from motion-induced disturbances and eliminating Abbe error. The motion architecture comprises a 3D air-bearing macro-motion stage coupled with a six-degree-of-freedom micro-motion stage； this macro-micro collaborative motion scheme improves measurement efficiency while enabling high-precision triggering and probing. For probing， either a high-precision 3D contact probe or a resonant probe is employed， with the resonant probe capable of using microspheres down to 70 μm in diameter. Major error sources were addressed by developing a spatial error-compensation model based on coordinate transformation and by applying error-separation techniques. Measurement performance was assessed in accordance with ISO 10360-2， yielding a maximum permissible error （MPE） better than ±（250 nm+3.6×10-6×L）. Experimental results demonstrate sub-micron 3D measurement accuracy and the ability to measure internal features as small as 100 μm. The proposed structural design and enabling technologies represent a significant advance for high-precision 3D metrology.关键词：micro-nano coordinate measuring machine;Abbe error-free;macro-micro motion system;precision measurement;error analysis11|3|0更新时间：2025-10-31
- 摘要：An input-shaping–based vibration suppression control method is proposed to mitigate mirror vibrations of a unimorph deformable mirror during operation. The mirror’s structural characteristics were analyzed to develop a dynamic model. The effects of input shapers on step-response characteristics （displacement overshoot and settling time） were then simulated. An experimental platform was constructed to validate the simulation results. Finally， the influence of the shaping technique on control performance was assessed via defocus wavefront reconstruction simulations. At a control frequency of 5 000 Hz， application of the input shaper reduced displacement overshoot from 32.1% to nearly zero and shortened settling time from 4.30 ms to 0.57 ms. In wavefront reconstruction simulations， overshoot decreased from 34% to less than 1%， while settling time improved from 3.95 ms to 0.58 ms， substantially enhancing both response speed and control accuracy. The designed input shaper is shown to effectively suppress vibrations of unimorph deformable mirrors.关键词：unimorph deformable mirror;input shaping;Vibration suppression;adaptive optics system10|4|0更新时间：2025-10-31
- 摘要：A linear-rotary piezoelectric motion platform based on the stick-slip driving principle was developed to address the requirements of robot-assisted target puncture. The platform integrates a two-degree-of-freedom flexure mechanism with an active locking flexure mechanism to achieve coupled linear-rotary motion while effectively suppressing backward displacement. Statics were derived using the compliance matrix method and a closed-loop serial flexure model， and structural parameters were optimized via a genetic algorithm. Static， modal， and transient response analyses were conducted using finite element analysis to determine the optimized amplification ratios and to validate the effectiveness of the active locking mechanism. Experimental results indicate amplification ratios of 2.865 and 1.568 for linear and rotary motions， respectively. The active locking mechanism reduced linear and rotary retraction displacements by 69.15% and 89.40%. The actuator attained a maximum linear velocity of 2 428.25 μm/s with a resolution of 0.569 μm. In the rotary direction， continuous 360° rotation was achieved with a maximum speed of 58 019 μrad/s and a resolution of 14.4 μrad. The actuator exhibited low coupling between motions and high bidirectional motion consistency （≥94%）， demonstrating suitability for target-puncture applications.关键词：piezoelectric actuator;stick-slip motion;precision actuation;compliance matrix9|7|0更新时间：2025-10-31
Micro/Nano Technology and Fine Mechanics
- 摘要：Accurate classification of visually similar reducer parts is essential for precise assembly. Existing visual classification methods struggle with highly similar parts due to limited discriminative features and low robustness to complex background interference， which can introduce errors in assembly. To address these challenges， a HiFuse-based Spatial Dual-Focus Synergy Network （SDFSN-HiFuse） is proposed for classification of reducer workpieces， targeting scenarios with large intra-class variance and small inter-class variance. A multi-branch spatially adaptive dilation-rate selection mechanism is introduced to enable automatic determination of appropriate receptive fields for deformed regions of workpieces. A two-stage geometric–local collaborative attention mechanism provides stepwise fine-grained guidance to features from each dilation branch， dynamically reweighting features and enhancing discrimination of salient regions via a coarse-to-fine refinement process. A deformable geometric graph is employed to model geometric topology flexibly， overcoming the constraints of traditional fixed grids. Following deformable convolution， a curvature gating mechanism preserves adaptive geometric deformation features， substantially improving responsiveness and representation accuracy on complex curved surfaces. On a custom dataset， SDFSN-HiFuse achieves a 3.57% absolute improvement in accuracy and a 2.99% increase in precision over the baseline， while meeting real-time requirements with a processing rate of 300.39 frame/s.关键词：reducer workpiece classification;deep learning;attention mechanism;multi-scale dilated convolution13|3|0更新时间：2025-10-31
- 摘要：A fast point-cloud registration method based on neighborhood distribution features is proposed to address the high computational cost of traditional high-dimensional feature extraction and the slow performance of dense registration algorithms that rely on coarse-fine two-step feature matching. First， three deep geometric features of neighboring points are defined， and a low-dimensional， multi-scale neighborhood distribution descriptor is introduced to substantially reduce feature-computation complexity while enhancing descriptor discriminability for efficient characterization of local point-cloud properties. Second， a rapid coarse-registration scheme is developed using the neighborhood distribution descriptor： feature points are selected according to the global undulation degree and neighborhood distribution direction； initial correspondences are established based on the neighborhood distribution descriptor； and Euclidean-distance constraints between point pairs are strengthened to remove incorrect matches， enabling efficient and accurate coarse alignment. Finally， to accelerate dense registration， the iterative closest point （ICP） algorithm is improved using a k-dimensional tree and voxel-grid downsampling， and a quadratic fine-registration strategy is employed to correct downsampling-induced errors， thereby further improving fine-registration accuracy and efficiency. Experiments on Stanford models and industrial part point clouds demonstrate that， compared with conventional feature-descriptor-based methods， the proposed approach increases registration accuracy by over 22% and reduces computation time by more than 43%， confirming its effectiveness， robustness， and practical applicability for rapid registration of object-surface point clouds acquired from different viewpoints.关键词：point cloud registration;neighborhood distribution feature;matched point pair optimization;iterative closest point;voxelized grid method11|3|0更新时间：2025-10-31
- 摘要：To address incomplete feature extraction， blurred boundaries， and omission of small targets in high-resolution remote sensing image segmentation， a detail-preserving segmentation method based on omnidirectional rectangular calibration， termed the Omnidirectional Rectangular Calibration Network （ORCNet）， is proposed. First， an Omnidirectional Rectangular Calibration State Space Module （ORSM） is introduced to enhance geometric adaptability and target retention via octagonal scanning and geometry-sensitive weight calibration. Next， a Complementary Filtering Hybrid Attention Fusion Module （CFHAF） is developed， integrating channel-， spatial-， and pixel-level attention mechanisms to enable adaptive multi-scale feature fusion and improved semantic discrimination. Finally， Dynamic Point Upsampling （DySample） is incorporated to refine boundary detail recovery. The model is trained with a Focal-Dice hybrid loss. On the Massachusetts Buildings dataset， the method achieves an F1 score of 84.64% and an mIoU of 77.07%. On the DeepGlobe Road dataset， an F1 score of 85.32% is obtained， representing a 3.51% improvement over RSMamba. Experimental results indicate that the proposed approach effectively addresses the three primary challenges in remote sensing segmentation， delivering high precision， robust performance， and strong potential for practical application.关键词：remote sensing image segmentation;rectangular self-calibration;state space model;multi-scale feature fusion;edge detail enhancement;dynamic upsampling9|2|0更新时间：2025-10-31
- 摘要：To mitigate the pronounced decline in vehicle detection performance caused by object blur and occlusion under adverse weather conditions， an enhanced YOLOv8-based vehicle detection algorithm， designated YOLO-CGT， is proposed. Tailored for vehicle-mounted camera imagery， the algorithm incorporates multiple enhancements to the YOLOv8 architecture to substantially improve detection robustness in challenging environments. Specifically， a multi-scale residual aggregation module replaces the original C2f module in the backbone network to increase exploitation of raw feature information and to alleviate gradient vanishing associated with greater network depth. A spatial aggregation module is incorporated to integrate global information extraction with local feature perception. Moreover， a lightweight dynamic detection head is developed to balance detection accuracy and computational efficiency. The conventional IoU metric is supplanted by the Inner-Minimum Points Distance Intersection over Union （Inner-MPDIoU） to reduce bounding-box overlap issues. Trained and validated on a vehicle dataset captured under complex weather conditions， the proposed method attains an average detection accuracy of 81.4%-an improvement of 6.3%-with 3.259×106 model parameters and a computational cost of 9.7 GFLOPs， demonstrating suitability for lightweight deployment while delivering substantial accuracy gains. These results provide a robust foundation for the safe and reliable operation of autonomous driving systems.关键词：intelligent driving;vehide detection;YOLOv8;complex weather;multi-scale features16|3|0更新时间：2025-10-31
- 摘要：To enhance the efficiency of container loading and unloading and realize automated operations， this study investigates monocular-vision-based automatic alignment between spreader and container， integrating deep learning and image-processing techniques. Monocular images of container hoisting conditions were analyzed with emphasis on the regional characteristics of corner components. To address the low pixel proportion of corner regions in high-resolution images， a two-stage "coarse positioning-fine segmentation" strategy was proposed. Based on the segmentation results， key feature points were detected， 2D-3D point correspondences were established， and pose estimation was performed using the Levenberg-Marquardt algorithm. Validation was conducted on an AUBO-i10 manipulator alignment platform in laboratory settings. Experimental results demonstrate that the mean average precision （mAP） for detection of container corner components exceeds 95% in both laboratory and real-scene environments. Mean intersection-over-union （mIoU） for corner segmentation reached 98.15% and 93.89%， respectively-improvements of 1.24% and 1.64% over the baseline SegFormer-B0-while model computational cost was reduced by approximately 23.2%. At a camera-to-corner distance of about 2 m， the aiming error of the alignment position was below 1.0 mm. Absolute translation errors on the X， Y， and Z axes were all below 5.0 mm， and absolute rotation error was below 0.5°. These results indicate that the proposed method achieves reliable accuracy and satisfies the requirements for automatic alignment of single-angle components.关键词：monocular vision;containers;automatic alignment;target recognition;pose estimation9|2|0更新时间：2025-10-31
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