最新刊期
卷 33 ， 期 22 ， 2025
- “在机械扫描激光雷达领域，专家建立了OSEDR数学模型，提出了基于扫描镜转速调控的探测器位置匹配方法，有效提升了探测距离和稳定性。”摘要：In mechanical scanning LiDAR， high-speed mirror rotation during long-distance detection induces an echo lag angle， causing image point deviation from the detector center and limiting the Optical Scanning Effective Detection Range （OSEDR）. A detector position matching method via scanning mirror speed control was proposed. First， a quantitative OSEDR model was established to describe the relationship among image point offset， scanning mirror rotation speed， target distance/azimuth， and optical system parameters. Then， based on the model， the rotation speed of the scanning mirror was adjusted， allowing conventional stable targets to serve as the feedback benchmark， replacing the specific long-distance targets required in traditional methods for detector position matching. Experimental results on a tower-based mirror scanning LiDAR system show that at the rated speed of 50 r/s， the maximum detection range increases from 1 020 m to 2 000 m， with an improvement of approximately 96%， meeting the theoretical detection range requirements. Moreover， the point clouds within this range are continuous and stable without obvious gaps. This method expands the selection of available feedback targets during debugging and provides a reliable and efficient approach to detector position matching in the assembly and alignment of mechanical scanning LiDAR systems.关键词：lidar;detector position matching;image point offset;rotational speed control;Optical Scanning Effective Detection Range （OSEDR）184|55|0更新时间：2026-01-15
- “在海洋环境监测领域，研究者提出了一种高灵敏度光纤盐度传感器，通过调整激光器采样间隔显著提升检测准确性。”摘要：Aiming at the problem of limited sensitivity of optical fiber salinity sensors， this paper proposed a high-sensitivity fiber optic Fabry-Perot （Fabry-Perot Interferometer， FPI） salinity sensor based on sparse sampling of a tunable laser. By controlling the sampling interval of the tunable laser to be close to the free spectral range （Free Spectral Range， FSR） of the FPI interferometric cavity， a vernier-like effect was formed in the spectrum. Under sparse sampling conditions， the interference signal generated adjustable envelope drift in the spectrum， which significantly amplified the refractive index perturbation caused by small salinity changes， thereby enabling high-sensitivity detection of salinity parameters. Experimental results demonstrate that the sensing system achieves a salinity sensitivity of up to 1.525 nm/‰ within the range of 20‰~35‰ salinity， which is approximately 9.24 times higher than that of a single FPI interferometer structure. Compared with the traditional vernier effect method based on a dual-cavity interferometer structure， the present sensor does not require any modification to the existing structure and can flexibly control system sensitivity and enhance detection accuracy simply by adjusting the sampling interval of the tunable laser source. The method is simple to implement， highly compatible， and has broad application prospects in the fields of high-performance fiber optic sensing and marine environment monitoring.关键词：tunable laser;sparse sampling;vernier effect;Fabry-Pérot;fiber optic salinity sensor71|66|0更新时间：2026-01-15
- “在森林生物量制图领域，专家提出了基于特征插值的多源数据融合方法，有效解决了星载激光雷达数据空间稀疏性问题，为大范围森林碳储量评估与生态系统监测提供方法参考。”摘要：Aiming at the problem of spatial sparsity of spaceborne lidar data， this paper proposed a multi-source data fusion method based on feature interpolation， which realized regional-scale forest aboveground biomass （AGB）. Firstly， the three-dimensional features were extracted from GEDI L2A/L2B and ICESat-2/ATL08 data； the data set of spot-scale feature variables was constructed by combining Sentinel-2 spectral feature variables and terrain factors. Then， the correlation analysis was carried out to eliminate the high-collinearity feature variables， and the three regression algorithms of CatBoost， RF and LightGBM were compared to identify the optimal model. Subsequently， based on CatBoost feature importance and SHAP analysis， key predictor variables were further identified. Finally， the key feature variables of LiDAR were interpolated to obtain continuous raster features， and then the forest AGB spatial mapping was realized by the optimal regression model. The validation results demonstrated that CatBoost performed best in spot-scale modeling （R²=0.88， RMSE=78.74 Mg/ha， rRMSE=20.93%）； the spatial mapping accuracy based on feature interpolation and multi-source data fusion is R2=0.82， RMSE=60.90 Mg/ha， and rRMSE=36.54%. Compared to regression mapping using optical remote sensing imagery alone， the rRMSE was reduced by approximately 34.7%. The feature interpolation strategy was used to spatially continuous the key structural variables of the laser spot and fuse them with high-resolution optical and topographic information， which can mitigate sparse laser-footprint sampling and the lack of vertical-structure information in optical images. It enhanced regional forest AGB estimation accuracy. And the method provides a valuable reference for large-scale forest carbon stock assessment and ecosystem monitoring.关键词：Global Ecosystem Dynamics Investigation（GEDI）;ICESat-2;lidar;optical remote sensing imagery;feature interpolation;aboveground biomass mapping86|164|0更新时间：2026-01-15
- “在三维激光扫描领域，专家提出了一种改进边界排斥的激光点云孔洞识别与修复方法，有效解决了内凹曲面或狭窄缝隙物体的点云孔洞问题，为激光点云孔洞修复提供了高效、可靠的解决方案。”摘要：Three-dimensional laser scanning efficiently constructs accurate 3D models. However， holes often form in point clouds when measuring objects with concave surfaces or narrow gaps due to light occlusion. These holes compromise model integrity and hinder subsequent applications. Traditional repair methods rely heavily on interpolation or surface fitting， which frequently results in non-uniform curvature and loss of detail in the repaired regions. This paper aimed to address these limitations. We proposed an improved hole identification and repair method based on boundary point repulsion. Firstly， the point cloud distribution was optimized using a Weighted Local Optimal Projection （WLOP） algorithm. Subsequently， hole boundaries were accurately identified by combining normal vector analysis with local density constraints. Maximum angle screening was employed to eliminate false boundary points， preventing misclassification of high-curvature areas as holes. Furthermore， a dynamic repulsion force field model was designed. A filling path was generated by leveraging the topological relationships of single-ring neighboring points， ensuring geometric consistency between the repaired and original point clouds. Finally， median filtering was applied to achieve smoothness consistency between the repaired hole area and the overall point cloud. Experimental results demonstrate the effectiveness of our method. It successfully identified and repaired various types of holes on multiple complex models. On the challenging Camel model， the repair rate reached 98.7%， and the point cloud density difference before and after repair was only 4.2%. The repair time was significantly shorter than that of LS-SVM， Poisson， PU-Net， and SeedFormer methods， being only 11.7% of the time required by SeedFormer. The proposed method exhibits significant advantages in preserving geometric structure and repairing details. It provides an efficient and reliable solution for laser point cloud hole repair.关键词：Laser point cloud;hole identification;point cloud repair;boundary point repulsion70|78|0更新时间：2026-01-15
Modern Applied Optics
- “在太阳能驱动界面蒸发技术领域，专家通过皮秒激光加工泡沫铝基界面蒸发体，得出最佳参数，制备的蒸发体具有出色的耐盐性能和长期稳定运行潜力，为可持续水净化技术实际应用提供新思路。”摘要：Regarding the solar-driven interfacial evaporation technology， in view of the huge challenges faced by interfacial evaporators in terms of preparation cost and long-term evaporation stability， research is carried out to prepare interfacial evaporators with high efficiency， low cost， and long - term stable operation， promote the practical application of sustainable water purification technology， and alleviate the global shortage of fresh water resources. First， the picosecond laser processing technology of aluminum foam-based interfacial evaporators was proposed to study the effects of different laser processing parameters such as single pulse energy， scanning speed， scanning interval， and number of repeated processing on the surface morphology and elements of aluminum foam interfacial evaporators. Secondly， the seawater desalination effects of evaporators under different parameters were compared to determine the optimal processing parameters. Finally， the water production performance was verified through outdoor seawater desalination experiments. The optimal processing parameters are as follows： scanning interval of 30 μm， single pulse energy of 380 μJ， scanning speed of 100 mm/s， and repeated processing for 1 time. When the light intensity is 1 kW/m²， the evaporation rate reaches 5.52 kg·m-2·h-1， and the water production in the real environment reaches 5.37 kg·m-2 over 8 hours. In 15 wt% high-concentration brine， the evaporation rate still remains at 4.03 kg·m-2·h-1. The concentrations of K⁺， Mg²⁺， Ca²⁺， B³⁺， and Na⁺ ions in the desalinated water are significantly reduced， meeting the drinking water standards of the World Health Organization （WHO）. This study investigates the fabrication of aluminum foam-based interfacial evaporators via picosecond laser processing， determining the optimal parameters. The prepared evaporators exhibit high efficiency， excellent salt resistance， and potential for long-term stable operation， providing new insights for the research and development of interfacial evaporators and promoting the practical application of sustainable water purification technologies.关键词：solar-driven interfacial evaporation technology;interfacial evaporator;picosecond laser;water production;salt resistance47|51|0更新时间：2026-01-15
- “在工业机器人领域，专家提出了基于可重构三维测量系统的误差补偿方法，有效提升了机器人定位精度，为多场景精度补偿提供新路径。”摘要：Aiming at the insufficient absolute positioning accuracy of industrial robots and the high cost， poor portability， and difficulty in adapting to multi-scene requirements of existing high-precision measurement systems for error acquisition， an error compensation method based on a reconfigurable three-dimensional measurement system was proposed to improve the robot positioning accuracy by integrating geometric and non-geometric error factors. First， the C-TrackTM|Elite high-precision measurement system and the lightweight binocular vision system were integrated to construct a bimodal reconfigurable measurement framework， and the high-precision coordinate solving was realized by the ellipse fitting and spatial position matching algorithms of the feature points and combining with the spherical fitting of the probes； then， the robot's position and distance error models were established separately， and the particle swarm algorithm was introduced to fit the Kriging experimental variability function automatically， and the error prediction equations were constructed to realize the robot's positional accuracy in the workspace； finally， the compensation was implemented and verified by comparison. The experiments show that： the position error compensation based on the high-precision system reduces the mean value of absolute error from 1.048 9 mm to 0.178 6 mm， and the accuracy is improved by 82.97%； the mean value of error after compensation of the lightweight system is reduced from 1.154 7 mm to 0.211 9 mm， and the accuracy is improved by 81.65%. It verifies the optimal balance between precision and cost of reconfigurable measurement system - high precision system is suitable for aerospace and other high value-added fields， lightweight system meets the precision needs of small and medium-sized enterprises， and provides a new technical path for multi-scenario precision compensation of industrial robots.关键词：reconfigurable;lightweight;industrial robot;Kriging;error prediction compensation103|139|0更新时间：2026-01-15
- “在精密位移测量领域，研究人员提出了一种新型微型光栅位移传感器，通过“W”形反射式折叠光路结构设计，实现了高量程-长度比和高精度测量，解决了狭小空间下的测量难题。”摘要：Displacement measurement at micro and nano scale holds paramount significance across domains such as mechanical precision machining， semiconductor manufacturing and industrial automation control. In comparison to traditional techniques， like potentiometer and laser interferometer， grating displacement measurement has the advantages of simple structure， high resolution and strong anti-electromagnetic interference ability， and has become the mainstream displacement measurement method in the field of precision measurement. However， most mature grating displacement sensors currently consist of measuring scales and reading heads， with large volumes and small ranges （low Range Length Ratio， RLR）. For instance， the RLR of Keyence's GT2-S1 is only 0.033， which cannot meet the application requirements in confined space scenarios. In response to that challenge， this paper proposed and implemented a novel micro grating displacement sensor， constructed a "W" - designing reflective folded optical path structure to compress the sensor's volume， integrated the sensor probe into the main body to achieve high RLR， and used the Particle Swarm Optimization （PSO） algorithm to identify and compensate for errors in photoelectric detection signals for high-precision measurement. After verification， the proposed sensor achieves a measuring range of 15 mm， an RLR of up to 0.6， and an error better than 1.1 μm when the volume is reduced to 7 mm×8.5 mm×25 mm. The sensor features compact size， high RLR， superior accuracy， and easy multiplexing， providing an effective solution for online gap and displacement measurement in confined spaces.关键词：absolute grating displacement sensor;miniaturization;displacement measurement;error compensation;gap measurement261|131|0更新时间：2026-01-15
- “在齿轮缺陷检测领域，研究者提出了一种基于多尺度特征融合与分块注意力的分割网络，有效提升了齿轮视觉特征的表征能力与细小缺陷的检测性能。”摘要：To address the limitations of traditional segmentation models in handling complex background interference and subtle defect regions of gears—particularly their insufficient feature-representation capability and poor robustness—this paper proposed a novel segmentation network based on multi-scale feature fusion and block-wise attention， aiming to enhance the representation of gear visual features and improve the detection performance of fine defects. First， a multi-scale feature-enhancement module replaced the standard downsampling blocks in the UNet encoder； it leveraged a parallel multi-branch convolutional structure to collaboratively extract multi-scale and multi-directional features， thereby enhancing the model’s perception of both local details and global context.Second， a block-wise feature-focusing module was introduced after downsampling； it employed a block-wise multi-head attention mechanism to independently analyze local regions， significantly improving sensitivity to minute defects and local texture variations.Finally， a weighted hybrid loss function was designed by combining Dice loss， binary cross-entropy （BCE） loss， and a gradient-difference constraint， effectively mitigating the class-imbalance issue and optimizing the quality of segmentation boundaries.Experimental results on both a self-constructed and public gear defect dataset demonstrate that the proposed method outperforms UNet and other state-of-the-art models in various gear defect segmentation tasks， achieving accuracy rates of 91.27% and 85.88%， respectively. The results validate the effectiveness and superiority of the proposed approach for precise detection and segmentation of surface defects in gears.关键词：surface defect detection;image segmentation;UNet;block-wise attention;gear81|92|0更新时间：2026-01-15
Micro/Nano Technology and Fine Mechanics
- “在三维激光扫描领域，专家提出了一种改进边界排斥的激光点云孔洞识别与修复方法，有效解决了内凹曲面或狭窄缝隙物体的点云孔洞问题，为激光点云孔洞修复提供了高效、可靠的解决方案。”摘要：Three-dimensional laser scanning efficiently constructs accurate 3D models. However， holes often form in point clouds when measuring objects with concave surfaces or narrow gaps due to light occlusion. These holes compromise model integrity and hinder subsequent applications. Traditional repair methods rely heavily on interpolation or surface fitting， which frequently results in non-uniform curvature and loss of detail in the repaired regions. This paper aimed to address these limitations. We proposed an improved hole identification and repair method based on boundary point repulsion. Firstly， the point cloud distribution was optimized using a Weighted Local Optimal Projection （WLOP） algorithm. Subsequently， hole boundaries were accurately identified by combining normal vector analysis with local density constraints. Maximum angle screening was employed to eliminate false boundary points， preventing misclassification of high-curvature areas as holes. Furthermore， a dynamic repulsion force field model was designed. A filling path was generated by leveraging the topological relationships of single-ring neighboring points， ensuring geometric consistency between the repaired and original point clouds. Finally， median filtering was applied to achieve smoothness consistency between the repaired hole area and the overall point cloud. Experimental results demonstrate the effectiveness of our method. It successfully identified and repaired various types of holes on multiple complex models. On the challenging Camel model， the repair rate reached 98.7%， and the point cloud density difference before and after repair was only 4.2%. The repair time was significantly shorter than that of LS-SVM， Poisson， PU-Net， and SeedFormer methods， being only 11.7% of the time required by SeedFormer. The proposed method exhibits significant advantages in preserving geometric structure and repairing details. It provides an efficient and reliable solution for laser point cloud hole repair.关键词：Laser point cloud;hole identification;point cloud repair;boundary point repulsion53|98|0更新时间：2026-01-15
- “科技媒体新闻记者报道：视觉Transformer结合多级聚合金字塔与多尺度注意力聚合模块，提出高精度边缘检测模型TFEdge，实验结果表明其优越性能。”摘要：The current mainstream edge detection method based on convolutional neural network has limitations in receptive field range and fine-grained edge perception. With the development of Vision Transformer， its global modeling ability and flexible information interaction mechanism bring new possibilities for edge detection tasks. To solve this issue， this paper proposed an encoder-decoder model named TFEdge， which combined Transformer， Multi-Level Aggregation Feature Pyramid （MLAFP）， and Multi-Scale Attention Aggregation （MSAA） modules for high-precision edge detection. The model introduced the Dilated Neighborhood Attention Transformer as the backbone network and extracted the global context information and local edge clues of the image through a multi-stage cascade design. Simultaneously， the Multi-Level Aggregation Feature Pyramid was designed to aggregate the deep and shallow features of each stage， endowing the shallow features with more abundant semantic features to suppress image noise and improve the detection ability of indistinct boundaries. Finally， the Multi-Scale Attention Aggregation module， based on an attention mechanism， was proposed to further enhance feature representation by aggregating the cross-scale spatial and channel attention information of feature maps.The experiment is evaluated on the BSDS500 and NYUDv2 datasets. The ODS and OIS F-scores of TFEdge on the BSDS500 are 0.857 and 0.874， respectively， while on the NYUDv2 they are 0.788 and 0.801， respectively. Compared with many existing methods， TFEdge shows superior edge detection performance in both quantitative and qualitative results.关键词：edge detection;transformer;attention mechanism;multi-level aggregation feature pyramid;multi-scale attention aggregation95|93|0更新时间：2026-01-15
-
Research and application of multi-focus image fusion based on CNN and Transformer architectures AI导读
“在图像融合领域，专家提出了一种多聚焦图像融合算法，有效提升了融合图像质量和下游任务效果，为实际检测任务中的快速性和精确性需求提供了解决方案。”摘要：To address the limitation that a single focused image could not simultaneously present complete scene information， this paper proposed an end-to-end multi-focus image-fusion algorithm aimed at enhancing fusion accuracy and practicality. A parallel encoder architecture that combined dense convolution and Transformer was constructed to effectively extract both high-frequency and low-frequency features. A spatial-attention mechanism was introduced to further enhance feature representation. In the fusion stage， a semantic-prior-guided cross-fusion strategy was designed to embed high-frequency details under the guidance of low-frequency information. This approach effectively mitigated the bias toward near-focus or far-focus regions seen in traditional methods and significantly improved contrast and detail preservation in the fused image. Compared with recent state-of-the-art methods and seven advanced image fusion algorithms on the Lytro， COCO and MFFW datasets， the proposed method demonstrates significant advantages across multiple metrics， achieving improvements of 2.7% in EN， 13.6% in PSNR， 7.9% in SSIM， 6.5% in MI， 1.6% in AG， and 3.7% in SF. In downstream tasks such as chip pin number recognition and chip center localization， the proposed method also achieves notable performance improvements， verifying its effectiveness and generalizability. The proposed network exhibits excellent performance in both fusion quality and downstream application tasks， meeting the requirements of fast and accurate multi-focus image fusion in practical scenarios.关键词：multi-focus image fusion;transformer;multi-head attention mechanism;chip recognition;chip inspection150|122|0更新时间：2026-01-15
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