最新刊期
卷 34 ， 期 7 ， 2026
- 摘要：Laser tracking measurement networks are widely employed for high-precision three-dimensional positioning， where accurate calibration of system parameters is essential for reliable results. Conventional network self-calibration methods often suffer from unstable geometric references， and the adjustment results are highly sensitive to initial values. An optimized adjustment method for laser tracking measurement networks is proposed. Oriented point models are introduced to provide multiple spatial references， enabling self-calibration of measurement station coordinates. During network adjustment， the coordinates of the oriented points are fixed， allowing joint optimization of all station coordinates under a unified geometric reference. The proposed method is based on the NASC model， in which rigid geometric constraints are imposed to enhance solution stability. Simulation experiments incorporating Gaussian noise in distance and point measurements demonstrate that stable solutions are achieved， with edge length errors controlled at the micrometer level. Practical validation was conducted using a one-meter standard rod. The proposed method was compared with the commercial software Spatial Analyzer （SA） through repeated measurements of two standard rod lengths within a 1.5 m×1.5 m measurement volume. Measurement errors obtained with the proposed method ranged from -0.3 μm to 4.8 μm， whereas those of SA were primarily distributed between 3.0 μm and 9.5 μm.The results indicate that the proposed method achieves higher accuracy， reduced error dispersion， and improved repeatability. This enhanced performance is attributed to the rigid geometric reference constraints of the NASC model， which effectively suppress uncertainty propagation during network adjustment. The method exhibits strong general applicability and supports high-precision spatial measurements， making it suitable for optical system assembly and other engineering tasks requiring micrometer-level accuracy. It provides a practical technical reference for high-precision measurement applications.关键词：pose measurement;Trilateration network adjustment;Geometric constraints;Network-Adjustment-based Self-Calibration model156|214|0更新时间：2026-04-10
- 摘要：To enhance the sensitivity and response speed of ammonia detection using laser absorption spectroscopy， optical feedback cavity-enhanced absorption spectroscopy （OF-CEAS） was employed. The characteristic absorption line of ammonia near 6 612.7 cm⁻¹ （1 512 nm） was selected as the target transition. A V-shaped optical resonant cavity composed of three high-reflectivity mirrors was constructed， with a single-arm length of 20 cm. The ring-down time of the empty cavity was measured to be 48.12 μs， corresponding to a finesse of approximately 100 000 and an equivalent absorption path length of 14.4 km. System stability was significantly improved through precise control of temperature and gas pressure. The optical setup was housed in a temperature-controlled chamber with a stability of ±0.005 °C. In addition， a self-developed high-precision pressure control system maintained the intracavity pressure at （30 000±1.5） Pa. Under these conditions， a detection limit as low as 12 parts per trillion （ppt） was achieved at an integration time of 168 s， corresponding to a minimum detectable absorption coefficient of 1.35×10⁻¹⁰ cm⁻¹. To mitigate response delays caused by adsorption of ammonia molecules on the inner surfaces of the gas cell and pipelines， a miniaturized cavity with a volume of 7.8 mL was implemented to reduce molecular residence. Furthermore， silane passivation coatings were applied to critical pipeline sections and the inner cavity surface to suppress adsorption effects， resulting in a reduction of the average system response time to 196 s.关键词：optical feedback cavity enhanced absorption spectroscopy;trace gas detection;high-precision temperature and pressure control;ammonia adsorption effect2|0|0更新时间：2026-04-10
- “介绍了其在精密制造领域的研究进展，相关专家开发了一种基于同轴光源照明的机器视觉自动测量系统，为大深径比微孔几何参数的高效、高精度测量提供了一种低成本技术方案。”摘要：To address the high equipment cost of conventional laser confocal microscopy and the limitations of traditional machine vision methods caused by uneven well-bottom illumination， image blurring， and difficulty in feature extraction， an automated machine vision measurement system based on coaxial illumination is proposed. The system comprises an industrial camera， a telecentric lens integrated with a coaxial light source， and a three-axis precision displacement stage. A hybrid algorithm combining autofocusing and feature recognition is employed as the core measurement strategy. Experimental results demonstrate stable and reliable system performance. For seven types of samples with aspect ratios ranging from 3.17 to 7.87， excellent measurement performance was achieved. The repeatability of aperture measurements ranges from 0.56 μm to 4.58 μm， while the repeatability of depth measurements reaches as low as 0.80 μm， with a minimum coefficient of variation of 0.09%. The deviation between the average measured values and nominal values is generally less than 1%. The proposed system enables efficient and high-precision measurement of geometric parameters of high-aspect-ratio micro-holes， providing a cost-effective solution for quality control in precision microstructure manufacturing.关键词：micropore measurement;aspect ratio;autofocus;Tenengrad gradient function;OTSU2|0|0更新时间：2026-04-10
- 摘要：Indirect microalgae density measurement methods based on short-range optical detection have increasingly attracted attention in the monitoring of marine microalgae cultivation， owing to their advantages of non-contact， non-invasive operation and the absence of reagent requirements. This review systematically summarizes prevalent cell measurement strategies， encompassing benchmark methods （manual counting and dry cell weight， DCW） and indirect optical approaches （optical sensing and image analysis）. The key characteristics and applicable conditions of each method are analyzed from four perspectives： detection range， accuracy， environmental interference， and equipment dependence. In light of existing limitations， the establishment of a standardized testing platform covering representative marine microalgae species， gradient cell densities， and diverse illumination and imaging conditions is proposed， together with a multidimensional， quantifiable， and reproducible evaluation framework. This work is expected to provide systematic references and technical guidance for the selection of microalgae bioprocess monitoring systems， algorithm optimization， and the development of novel sensing paradigms.关键词：optical measurement;marine microalgae;density monitoring;image analysis2|0|0更新时间：2026-04-10
- 摘要：In the fabrication of Mo-Si multilayers， achieving smooth and sharp interfaces is critical for realizing high reflectivity in extreme ultraviolet （EUV） light， as atomic intermixing between adjacent Mo and Si layers along with microscopic interfacial fluctuations can significantly degrade EUV reflectivity. To address this challenge， we propose a process portfolio that combines angular deposition with flood ion beam etching to enhance interface quality. By utilizing stage modifications in dual ion beam sputtering， the incident angle of the sputtered atom flux during deposition and the parameters of the auxiliary ion beam for polishing can be precisely controlled to suppress intermixing and interfacial fluctuations. Experimental results show that the thickness of the intermixing layer is reduced to 0.6 nm， and interfacial roughness is suppressed to 0.2 nm using this approach. This method fundamentally improves the interfacial quality of Mo-Si multilayers and offers a practical solution for the fabrication of high-reflectance EUV optics.关键词：Mo-Si multilayer reflective film;EUV lithography;dual ion beam sputtering deposition;oblique angle deposition;glancing angle etching2|0|0更新时间：2026-04-10
Modern Applied Optics
- “针对复杂扰动环境下激光跟踪测量系统动态响应滞后和控制精度下降问题，专家提出基于径向基函数神经网络优化的线性自抗扰控制方法，为提升系统动态性能与控制精度提供解决方案。”摘要：Laser tracking measurement systems frequently operate under complex disturbances， including nonlinear friction and inter-axis coupling， which induce response delays and degrade control accuracy. To address these issues， a radial basis function neural network-optimized linear active disturbance rejection control （RBF-LADRC） method is proposed. In this approach， external disturbances and model uncertainties are aggregated as a total disturbance within the LADRC framework. A linear extended state observer is employed to estimate this total disturbance in real time and compensate for it online. Furthermore， a radial basis function neural network is introduced to identify the Jacobian of the controlled plant online. Based on the identified Jacobian， a gradient descent algorithm is constructed to enable adaptive updating of controller gains. The stability and parameter convergence of the closed-loop system are established using discrete Lyapunov theory. Experimental validation on a laser tracking measurement system demonstrates that， compared with conventional LADRC， the proposed method enables dynamic adaptation of controller parameters to varying operating conditions. The system bandwidth is increased by approximately 12%， the settling time is reduced by about 32%， and the root mean square tracking error of the laser trajectory is decreased by approximately 16%. The proposed RBF-LADRC method does not rely on an accurate system model and effectively enhances both dynamic performance and control precision， indicating strong potential for engineering applications.关键词：laser tracking measurement;disturbance rejection;linear active disturbance rejection control;radial basis function neural network2|0|0更新时间：2026-04-10
- 摘要：To mitigate measurement errors induced by illumination variations in precision image measurement， an error compensation model is proposed based on multidimensional grayscale features extracted via the Segment Anything Model （SAM） and fitted using a Whale Optimization Algorithm-optimized Radial Basis Function （WOA-RBF） neural network. A mathematical model describing illumination-induced edge shift is established to characterize the nonlinear effects of light intensity and surface scattering properties on measurement accuracy. Leveraging SAM's zero-shot segmentation capability， average grayscale values from heterogeneous material regions are automatically extracted as multidimensional feature inputs to represent complex image characteristics. The WOA is employed to optimize the parameters of the RBF neural network， enabling accurate compensation of edge shift errors. Comparative experiments on chromium-zirconium-copper fixture measurements， benchmarked against one-dimensional linear fitting， GA-LSSVM， and SVR methods， demonstrate that the proposed model achieves an RMSE of 2.07 μm， an MAE of 1.73 μm， and an R² of 0.99 （with the Zernike moment sub-pixel algorithm as a representative case）. Consistent accuracy and strong robustness are observed across various sub-pixel edge detection algorithms， indicating that the proposed approach provides an effective solution for illumination-induced errors in precision image measurement.关键词：computer vision;edge detection;error compensation;segment anything model;whale optimization algorithm;radial basis function neural network2|0|0更新时间：2026-04-10
-
Robot precision calibration based on fusion of physical prior and progressive decoupling network AI导读
“介绍了其在机械臂标定领域的研究进展，相关专家提出融合物理先验与渐进式解耦残差网络的标定方法，为解决多自由度机械臂标定难题提供解决方案。”摘要：To overcome the limitations of traditional geometric calibration in compensating for non-geometric errors， as well as the poor interpretability and susceptibility to gradient competition in multi-dimensional heterogeneous error fields inherent in purely data-driven black-box models， an absolute accuracy calibration method integrating physical priors with a progressively decoupled residual network is proposed.First， a differentiable kinematic grey-box model based on Denavit–Hartenberg （DH） parameters is constructed as an explicit physical framework to compute the baseline theoretical pose. Second， high-dimensional sine–cosine encodings and second-order multiplicative combinatorial features are introduced to enhance the representation of periodic nonlinear errors. A dual-branch residual network is then employed to independently predict position and orientation residuals， incorporating a differentiable singular value decomposition （SVD） orthogonalization layer to strictly enforce SO（3） manifold constraints.Furthermore， a stage-wise parameter freezing strategy is designed to enable progressive decoupled training， effectively mitigating optimization conflicts arising from the differing dimensionalities of position and orientation. Experimental results on a Staubli TX2-90L demonstrate that the average position error is reduced from 0.377 mm to 0.047 mm. Compared with support vector regression （SVR） and backpropagation （BP） methods， positioning accuracy is improved by 26.3% and 49.9%， respectively. The proposed method achieves a favorable balance between high precision and interpretability， indicating substantial potential for engineering applications such as in situ bioprinting.关键词：multi-degree-of-freedom robotic arm;kinematic calibration;non-geometric error;residual network2|0|0更新时间：2026-04-10
Micro\/Nano Technology and Fine Mechanics
- “针对复杂交通场景下光照变化与跨模态干扰导致自动驾驶感知能力退化的问题，提出一种基于可靠性自适应引导的红外与可见光图像融合网络。该方法通过构建像素级可靠性度量机制，联合建模结构一致性与强度异常以动态评估信源可信度，在全局层采用“可信注入”策略校正强度分布，在细节层利用自适应引导滤波实现显著目标与纹理的竞争增强，并结合多约束损失函数协同优化。在M3FD与RoadScene数据集上的实验结果表明，相较于DWT，GTF，U2Fusion及Umcfuse等主流算法，本文方法在信息熵、标准差、空间频率、平均梯度、互信息、融合质量、边缘强度及视觉信息保真度等分别平均提高了1.51%，16.56%，42.36%，52.24%，38.28%，80.51%，21.4%和17.6%；在下游目标检测任务中平均精确率达91.4%，优于其他融合方法。该方法有效抑制了伪影与噪声，具备优异的场景泛化性与稳定性，能显著提升自动驾驶系统的环境感知精度。”摘要：To mitigate perception degradation in autonomous driving caused by illumination variations and cross-modal interference， an infrared-visible image fusion network with reliability-adaptive guidance is proposed. A pixel-level reliability estimation mechanism is constructed by jointly modeling structural consistency and intensity anomalies， enabling dynamic assessment of source credibility. A “trusted injection” strategy is introduced to correct the global intensity distribution， while adaptive guided filtering enhances the competition between salient objects and texture details in the detail layer； the process is optimized באמצעות a multi-constrained loss function. Experiments on the M3FD and RoadScene datasets demonstrate that， compared with DWT， GTF， U2Fusion， and Umcfuse， the proposed method improves EN， SD， SF， AG， MI， Qabf， EI， and VIFF by 1.51%， 16.56%， 42.36%， 52.24%， 38.28%， 80.51%， 21.4%， and 17.6%， respectively. In downstream target detection tasks， an average accuracy of 91.4% is achieved， surpassing existing fusion methods. The proposed approach effectively suppresses artifacts and noise， exhibits strong scene generalization and robustness， and significantly enhances environmental perception accuracy in autonomous driving systems.关键词：image fusion;infrared and visible;reliability-adaptive guidance;cross-modal structural consistency;trusted injection;autonomous driving perception2|0|0更新时间：2026-04-10
- 摘要：Detection of material mixing uniformity is critical for enabling online quality monitoring and process optimization. This study addresses the degradation of uniformity evaluation caused by defocus blur in hyperspectral imaging （HSI）. A physics-constrained self-supervised learning framework for unpaired hyperspectral image deblurring （PC-SSL-HSI） is proposed. A Uformer-based architecture incorporating the SimAM attention mechanism is employed as the deblurring network， while adversarial training is introduced to align deblurred outputs with clear images in the feature space. In addition， a blur kernel prediction module is designed based on a classical degradation model to construct pseudo-sample pairs， enabling self-supervised learning that guides the network to emphasize local detail restoration in hyperspectral images.Experimental results demonstrate that the proposed method effectively enhances image detail， suppresses artifacts， and improves the accuracy of material mixing uniformity evaluation. On a simulated dataset， the peak signal-to-noise ratio （PSNR） reaches 34.970 and the structural similarity index （SSIM） reaches 0.900， with concentration prediction errors ranging from 0.022 8 to 0.031 2. Furthermore， hyperspectral imaging experiments for material mixing uniformity indicate that the proposed method outperforms comparative approaches in metrics such as Kullback–Leibler divergence and coefficient of variation， highlighting its strong potential for engineering applications.关键词：hyperspectral image;defocus deblur;mixing uniformity;self-supervised learning;physics-constrained2|0|0更新时间：2026-04-10
-
Dual-scale tableau detail information-guided full-reference objective assessment on image quality AI导读
“在工业视觉领域，专家提出一种双尺度画面细节信息引导的全参考图像质量评价模型，为图像质量评价提供新方法。”摘要：In industrial vision systems， full-reference image quality assessment （FR-IQA） has emerged as a major research focus. To address this problem， a novel FR-IQA model， termed a dual-scale tableau detail-guided model， is proposed. First， the test image is transformed into a specific color space to decouple luminance and chrominance channels. Subsequently， a luminance gradient similarity component is constructed based on gradient interactions among the reference image， the distorted image， and their fused representation. This component is further integrated with a chrominance orientation consistency analysis to derive a color similarity measure. Meanwhile， a dual-scale tableau detail accumulation component is developed by fusing spectral residual， edge， and adaptive contrast features of the luminance channel， enabling effective characterization of cumulative detail information. Finally， the standard deviation features of these components are weighted and aggregated through feature coefficients to produce the overall quality score.The reliability of the proposed model is validated on the LIVE， CSIQ， TID2008， TID2013， and KADID-10K databases using four evaluation metrics： Spearman rank-order correlation coefficient （SROCC）， Pearson linear correlation coefficient （PLCC）， Kendall rank-order correlation coefficient （KROCC）， and root mean square error （RMSE）. Experimental results indicate that the PLCC ranges from 0.876 8 on KADID-10K to 0.967 8 on LIVE， while the SROCC varies from 0.864 8 on TID2013 to 0.961 0 on CSIQ. The proposed model demonstrates superior computational efficiency compared with state-of-the-art and deep learning-based FR-IQA methods， while maintaining robust and generalizable predictive performance.关键词：image quality assessment;full reference;dual-scale;tableau detail information2|0|0更新时间：2026-04-10
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