检索结果-内蒙古大学图书馆

Event-Based Stereo Depth Estimation With Motion Guidance and Left–Right Consistency

IEEE Transactions on Instrumentation and Measurement 2025年 74卷

作者： Jiang, Junjie Zhuang, Hao Huang, Xinjie Kong, Delei Fang, Zheng Northeastern University Faculty of Robot Science and Engineering Shenyang110819 China Hunan University School of Robotics Changsha410082 China

Depth estimation, which directly captures the structure of observable environmental surfaces, plays a critical role in vision-based applications such as measurement, mapping, autonomous driving, and robot navigation. Specifically, event camera-based stereo depth estimation provides a novel solution to challenging conditions like rapid motion and extreme illumination variations. Currently, deep learning has become the dominant approach for event camera-based stereo depth estimation. However, these methods fail to fully exploit temporal cues in the event stream, resulting in suboptimal event representation clarity. Furthermore, there is room for further reduction in pixel shifts in feature maps before constructing the cost volume. This article presents a novel event-based stereo depth estimation method with motion guidance and left-right consistency. First, an edge-aware aggregation (EAA) module is proposed, which integrates event frames with motion confidence maps to generate a novel high-definition event representation. Then, a motion-guided attention (MGA) module is introduced, which leverages deformable transformer encoders guided by motion confidence maps to refine edge precision in feature maps. Finally, a census left-right consistency loss function is designed to enhance the left-right consistency of stereo event representation. Experiments conducted in challenging real-world driving scenarios demonstrate that the proposed method outperforms state-of-the-art (SOTA) methods in terms of mean absolute error (MAE) and root mean square error (RMSE) metrics. © 1963-2012 IEEE.

关键词： Stereo image processing

来源：评论

学校读者我要写书评

暂无评论

A Space Non-Cooperative Object Active Recognition Method Based on Zoom Area Extract

A Space Non-Cooperative Object Active Recognition Method Bas...

引用

Chinese Automation Congress (CAC)

作者： Pengfei Zhao Feng Zhu Haibo Sun Jianyu Wang Qun Wang Yangyang Li Key Laboratory of Opto-Electronic Information Processing Chinese Academy of Sciences Shenyang China Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China University of Chinese Academy of Sciences Beijing China Shanghai Institute of Microsystem and Information Technolohy Chinese Academy of Sciences Shanghai China Faculty of Robot Science and Engineering Northeastern University Shenyang China

ISBN: (数字)9798350368604

ISBN: (纸本)9798350368611

Space non-cooperative object detection is a key technology for on-orbit mission. At present, the space non-cooperative object detection method based on machine learning improves the detection accuracy of the target from different perspectives. However, non-cooperative object lacks prior information, and space tasks have a certain urgency, which underscores the need for efficient implementation of detection methods. Active filtering of image information is a challenge to improve the performance of space non-cooperative object detection. To address the challenges in space applications, we designed a space non-cooperative object detection method based on active zoom area extraction. This method consists of two modules: an active zoom area extraction module, which combines reinforcement learning and feature weight heatmaps, and an object detection module utilizing ResNet. The active zoom area extraction module down-samples the original image and employs the strategy network to actively select the zoom area with the highest probability of containing the object. It then optimizes this zoom area using feature weighting to obtain the final zoom area. The object detection module detects the zoom area and the downsampled image. According to the experimental results of two public datasets, this method can actively extract the zoom area without reducing the detection accuracy as much as possible, effectively reduce the processed image information, and improve the computational efficiency.

关键词： Accuracy Image recognition Filters Space missions Object detection Reinforcement learning Feature extraction Computational efficiency Data mining Space heating

来源：评论

学校读者我要写书评

暂无评论

Maximizing RbGeI3 perovskite solar cell efficiency through advanced TiO2/graphene nanocomposite electron transport layer

引用

Optik 2025年 320卷

作者： Bin Abdul Ghani, Ibtasam Khalid, Misha Yan, Haoran El Arfaoui, Youssef Nawaz, Bilal Wang, Jianxin Key Laboratory of Advance Technologies of Materials Ministry of Education College of Medicine School of Materials Science and Engineering Southwest Jiaotong University Chengdu610031 China Institute of Physics Faculty of Physics Astronomy and Informatics Nicolaus Copernicus University Grudziadzka 5 Torun87100 Poland College of Chemistry and Materials Engineering Anhui Science and Technology University Bengbu233030 China Department of Physics Faculty of Science Ibn Zohr University Agadir80000 Morocco State Key Laboratory of Rolling and Automation Northeastern University Liaoning Shenyang110819 China

The practical commercialization of lead-based perovskite solar cells (PSCs) has been hindered by lead toxicity, despite their low production cost and high efficiency. Recently, researchers have intensified to reduce the toxicity of PSCs by developing alternative lead-free compositions. This study aims to optimize lead-free RbGeI3-based PSC through SCAPS-1D tool, by utilizing pristine TiO2 and TiO2/graphene (Gr) nanocomposite at different concentrations (0.5 %, 1 %, 1.5 %) as electron transport layer (ETL) and NiO as hole transport layer (HTL). The impact of TiO2+Gr nanocomposite ETLs on PSC performance was examined, focusing on layer thickness, doping concentrations, defect densities in the perovskite layer and at the interfaces, and temperature effect. When pristine TiO2 was used as the ETL, the device exhibited an open circuit voltage (VOC=0.9126 V), current density (JSC=31.59 mAcm−2), and power conversion efficiency (PCE=24.91 %). However, with the optimal TiO2+Gr(0.5 %) ETL, improved carrier transport and reduced interface defects boosted VOC=1.0622 V, and JSC=31.98 mAcm−2. This led to a 30.14 % enhancement in cell efficiency, resulting in 19 % increase in PCE compared to the undoped TiO2-ETL. © 2024 Elsevier GmbH

关键词： Perovskite

来源：评论

学校读者我要写书评

暂无评论

Ultra-Sensitive Fiber Optic Temperature Sensor Based on Vernier Effect in Cascading Sagnac Loops Integrated with Pmfs

SSRN

引用

SSRN 2024年

作者： Liu, Qiang Zhang, Yu Xu, Tian-Shuai Wu, Qiang Fu, Yongqing Ng, Wai Pang Fu, Yongqing Binns, Richard School of Control Engineering Hebei Key Laboratory of Micro-Nano Precision Optical Sensing and Measurement Technology Northeastern University at Qinhuangdao Qinhuangdao066004 China College of Information Science and Engineering Northeastern University Shenyang110819 China Faculty of Engineering and Environment Northumbria University Newcastle upon TyneNE1 8ST United Kingdom Northumbria University United Kingdom

An ultra-sensitive temperature sensor was proposed and demonstrated by inserting polarization maintaing fibers (PMFs) into cascading Sagnac fiber loops. Effect of the length of PMFs in an individual Sagnac loop on the temperature sensitivity of the sensor was firstly studied. As the PMF length reduces, the sensitivity increases. In the PMFs cascade Sagnac loops, the closer the lengths of the two sections of PMFs embedded in the Sagnac loops, the wider the free spectral range, resulting stronger Vernier effect and thus higher sensitivity. The temperature sensitivity of the cascade Sagnac loops is ~-44.28 nm/°C, a 27-fold increase than that of an individual Sagnac loop (~-1.64 nm/°C). This newly developed temperature sensor has the merits of simple structure, high sensitivity and easy operation, and good competitiveness in the sensing field. © 2024, The Authors. All rights reserved.

关键词： Polarization-maintaining fiber

来源：评论

学校读者我要写书评

暂无评论

Electroactive Marine Actinomycete Nocardiopsis Dassonvillei Accelerates the Corrosion of Aluminum Alloy

SSRN

引用

SSRN 2024年

作者： Gao, Yu Narenkumar, Jayaraman Wang, Donglei Liu, Qiance Fan, Jiaxin Wang, Fuhui Jiang, Guangming Xu, Dake Moradi, Masoumeh Shenyang National Laboratory for Materials Science Northeastern University Shenyang110819 China SDU Life Cycle Engineering Department of Green Technology Faculty of Engineering University of Southern Denmark Odense5230 Denmark School of Civil Mining and Environmental Engineering University of Wollongong WollongongNSW2522 Australia Northeastern University United States Key Laboratory for Anisotropy and Texture of Materials Ministry of Education China

Microbial corrosion based on extracellular electron transfer (EET) has been widely studied on iron-based metals, however, it is still scarce for aluminum alloys. In this study, the corrosion behavior of a novel isolated marine actinomycete, Nocardiopsis dassonvillei, against AA5083 was systematically evaluated. The results proved that N. dassonvillei biofilm accelerated AA5083 corrosion by increasing the dissolution of oxide layers through chloride accumulation, preventing the re-passivation through oxygen depletion. On the metal substratum with the defected passive film, N. dassonvillei accelerated the Al dissolution by facilitating inward EET with phenazine as an electron shuttle between metal surface and bacterial cells. © 2024, The Authors. All rights reserved.

关键词： Aluminum alloys

来源：评论

学校读者我要写书评

暂无评论

COFH——An improved feature descriptor for RGB information based on OVFH

COFH——An improved feature descriptor for RGB information b...

引用

第33届中国控制与决策会议

作者： Yang Jiang Weiping Li Xingmao Wu Xiao Chen Faculty of Robotic Science and Engineering Northeastern University

In this paper,we propose a novel point cloud feature descriptor for object recognition based on 3 D RGB point *** to its strong recognition power,OVFH(Orthogonal Viewpoint Feature Histogram) feature descriptor is widely used in real-time 3 D point cloud object recognition task,especially when dealing with the object of symmetric placement of mirror image,and has achieved good ***,this descriptor does not take the color information of the point cloud into consideration,which is an essential information for object *** solve this problem,we propose a novel feature descriptor named COFH(Color-OVFH).It consists of two parts,which are viewpoint component and extended color shape *** extended color shape feature integrates the surface shape feature and color *** performance of COFH descriptor is validated on the common *** experimental result shows that COFH significantly improved the recognition ability in RGB point clouds with similar shapes but large color *** can be effectively applied to object recognition with RGB information.

关键词： Global Feature Descriptor Object Recognition RGB Point Cloud Viewpoint Feature Histogram

来源：评论

学校读者我要写书评

暂无评论

Reducing-Over-Time Tree for Event-based Data

Reducing-Over-Time Tree for Event-based Data

引用

International Conference on Pattern Recognition

作者： Shane Harrigan Sonya Coleman Dermot Kerr Pratheepan Yogarajah Zheng Fang Chengdong Wu School of Computing Engineering and Intelligent Systems Ulster University Northern Ireland UK Faculty of Robot Science and Engineering Northeastern University Liaoning China

This paper presents a novel Reducing-Over-Time (ROT) binary tree structure for event-based vision data and subtypes of the tree structure. A framework is presented using ROT, that takes advantage of the self-balancing and self-pruning nature of the tree structure to extract spatial-temporal information. The ROT framework is paired with an established motion classification technique and performance is evaluated against other state-of-the-art techniques using four datasets. Additionally, the ROT framework as a processing platform is compared with other event-based vision processing platforms in terms of memory usage and is found to be one of the most memory efficient platforms available.

关键词： Matched filters Image edge detection Memory management Binary trees Data mining Corner detection

来源：评论

学校读者我要写书评

暂无评论

Twin Delayed Deep Deterministic Policy Gradient Based Energy Management Strategy for Fuel Cell/Battery/Ultracapacitor Hybrid Electric Vehicles Considering Unknown Terrain Information

SSRN

引用

SSRN 2023年

作者： Tao, Fazhan Fu, Zhigao Gong, Huixian Ji, Baofeng Zhou, Yao School of Information Engineering Henan University of Science and Technology Luoyang China Longmen Laboratory Luoyang China Faculty of Science and Technology University of Macau China Henan Key Laboratory of Robot and Intelligent Systems Henan University of Science and Technology Luoyang China

For fuel cell/battery/ultracapacitor hybrid electric vehicles (FCHEV), the complex topology and unknown terrain information pose challenges to fuel cell efficiency, power sources lifespan, and fuel economy. In this paper, a hierarchical energy management strategy (EMS) considering unknown terrain information is constructed to optimize the demand power allocation of FCHEV based on twin delayed deep deterministic policy gradient (TD3). Firstly, to improve the fuel economy of FCHEV, an adaptive fuzzy filter-based upper-level frequency separation is designed considering negative power. Secondly, to achieve the minimum fuel consumption, a TD3-based lower-level EMS is designed, where the unknown terrain information is predicted through Multi-LSTM network and considered in the TD3 framework. The optimal state of charge obtained by dynamic programming (DP) based on the demand power and predicted terrain information, is considered in the reward function of TD3, as well as the optimal experience solved by DP based on historical road data is integrated into the designed hybrid experience replay. Finally, the simulation results under SC03 and test driving cycles indicate that compared to traditional TD3-based EMS, the fuel economy is increased by 33.3% and 27.5%, respectively, and can reach the benchmark level of 80% and 80.9% of DP, respectively. © 2023, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Pest Rat Detection and Capture System in Barn

Pest Rat Detection and Capture System in Barn

引用

Intelligent robotics and Automatic Control (IRAC), International Conference on

作者： Zizheng Rui Yuhang He Wenxuan Yang Hongxuan Xu Yadong Xie Jiayi Wang College of Ocean Engineering Harbin Institute of Technology Harbin China School of Computer Science Northwestern Polytechnical University Xi'an China School of Materials Science and Engineering Harbin Institute of Technology Harbin China School of Computer and Communication Engineering Northeastern University at Qinhuangdao Qinhuangdao China Faculty of Computing Harbin Institute of Technology Harbin China

ISBN: (数字)9798350389807

ISBN: (纸本)9798350389814

With the development of agricultural automation and intelligence, pest rat monitoring in barns becomes more and more important. The traditional rat detection method has problems such as low precision and poor efficiency, which is difficult to meet the practical needs. In this paper, an improved method for rat detection in barn is proposed based on YOLOv5 algorithm. First, in order to improve the detection accuracy, a multi-scale feature fusion strategy is adopted, which improves the network structure in order to better capture the features of rats at different scales; second, data enhancement techniques, including image rotation, scaling, and color transformations, are used to enhance the robustness and generalization ability of the model. Finally, in order to improve the real-time detection capability, lightweight network design and quantization techniques are used to reduce the computational and parameter numbers of the model, thus achieving faster detection. The experimental results show that the improved YOLOv5 model has a greater improvement in both accuracy and detection speed compared to the original YOLOv5 model on the rat detection task. The method proposed in this paper has greater potential for application in agricultural surveillance and provides technical support for automated pest rat control.

关键词： YOLO Accuracy Quantization (signal) Computational modeling Surveillance Memory management Rats Feature extraction Transformers Robustness

来源：评论

学校读者我要写书评

暂无评论

PointStack: A Point Cloud Processing Network with Enhanced Global Feature Aggregation

PointStack: A Point Cloud Processing Network with Enhanced G...

引用

Software engineering and Computer science (CSECS), International Conference on

作者： Yongliang Tao Jintao Wang Pengchao Li Mingmin Liu Zhenjun Du Yang Liu Chengpeng Li Faculty of Robot Science and Engineering Northeastern university Shenyang China Central Research Academy ShanDong SIASUN Industrial Software Research Institute Co. Ltd Shandong China Central Research Academy SIASUN Robot & Automation Co. Ltd Shenyang China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China

Point cloud data, characterized by its rich geometric information, dense data representation, and flexible topological structure, finds extensive application in fields such as computer graphics, computer vision, and robotics perception. The rapid advancement and upgrade of three-dimensional point cloud acquisition devices have made the acquisition of point cloud data increasingly convenient. This, in turn, has spurred an urgent need for research on point cloud data processing. The irregularity and disorderliness of point clouds necessitate the design of network models that differ from those used for image processing. Presently, most network models for point cloud processing focus primarily on the design for local feature extraction, even though the effectiveness of these improvements has become somewhat marginal. In response to this, a stacked diffusion network model designed to capture global context has been proposed. This model excels at capturing global information and can effectively integrate it with local information. Extensive experiments conducted on standard benchmark datasets such as ModelNet40 and ShapeNetPart indicate that this model, in comparison to DGCNN, improves point cloud classification by 1.2%, achieving an Overall Accuracy (OA) of 94.1%. Furthermore, it enhances point cloud part segmentation by 1.3%, reaching an mIOU evaluation metric of 86.5%. When compared to other advanced methods, this model also exhibits noticeable improvements in point cloud classification and part segmentation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：