检索结果-内蒙古大学图书馆

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Juncheng Mu Lin Bie Shaoyi Du Yue Gao BNRist THUIBCS School of Software Tsinghua University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Point cloud registration is still a challenging and open problem. For example, when the overlap between two point clouds is extremely low, geo-only features may be not suf-ficient. Therefore, it is important to further explore how to utilize color data in this task. Under such circumstances, we propose ColorPCR for color point cloud registration with multi-stage geometric-color fusion. We design a Hier-archical Color Enhanced Feature Extraction module to ex-tract multi-level geometric-color features, and a GeoColor Superpoint Matching Module to encode transformation-invariant geo-color global context for robust patch corre-spondences. In this way, both geometric and color data can be used, thus leading to robust performance even under extremely challenging scenarios, such as low overlap between two point clouds. To evaluate the performance of our method, we colorize 3DMatch/3DLoMatch datasets as Color3DMatch/Color3DLoMatch and evaluations on these datasets demonstrate the effectiveness of our proposed method. Our method achieves state-of-the-art registration recall of 97.5%/88.9% on them.

关键词： Point cloud compression Computer vision Accuracy Image color analysis Feature extraction Pattern recognition Task analysis

来源：评论

学校读者我要写书评

暂无评论

Internet of Things (IoT) and Machine Learning (ML) in Cyber-Security 2

Internet of Things (IoT) and Machine Learning (ML) in Cyber-...

引用

2nd IEEE International Conference on Artificial Intelligence of Things and Crowdsensing, AIoTCs 2022

作者： Abdelrahim, Malik Al-Turjman, Fadi Research Center for Ai and IoT Near East University Artificial Intelligence Engineering Department Nicosia Cyprus Artificial Intelligence Engineering Department Ai and Robotics Institute Near East University Mersin10 Turkey

ISBN: (纸本)9798350334104

With the technological advancement, the world is going through, at such a high pace, new challenges arise that might require new approaches. ML is, without a doubt, on the lead of technology, as it is evolving with minimal human intervention. However, new models of ML are crucial to preventing hackers from manipulating data. It is true that with new models, new gates for attacks are opened, which is why ML security must follow the same pace in evolving. In this paper, a brief introduction to ML and ML security will be discussed to give an insight on the topic, before going through a more specific point;the challenges faced by ML security. By introducing these terms, the reader will come to grasp the importance of ML security, and the huge role it plays when it comes to protecting data against hackers. However, this by no means indicates that it is flawless;on the contrary, there are flaws in the methods used in our present day, which is another topic this paper will give an insight on. Our main focus in this paper will be supervised learning due to its application in security. Algorithms used in ML models is the main gate that hackers use to breach the system, for this precise reason, engineers and experts are challenged to come with new algorithms, as we will conclude this work. © 2022 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement

Salience DETR: Enhancing Detection Transformer with Hierarch...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Xiuquan Hou Meiqin Liu Senlin Zhang Ping Wei Badong Chen National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an China College of Electrical Engineering Zhejiang University Hangzhou China

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

DETR-like methods have significantly increased detection performance in an end-to-end manner. The main-stream two-stage frameworks of them perform dense self-attention and select a fraction of queries for sparse cross-attention, which is proven effective for improving performance but also introduces a heavy computational burden and high dependence on stable query selection. This paper demonstrates that suboptimal two-stage selection strategies result in scale bias and redundancy due to the mismatch between selected queries and objects in two-stage initial-ization. To address these issues, we propose hierarchical salience filtering refinement, which performs transformer encoding only on filtered discriminative queries, for a bet-ter trade-off between computational efficiency and precision. The filtering process overcomes scale bias through a novel scale-independent salience supervision. To com-pensate for the semantic misalignment among queries, we introduce elaborate query refinement modules for stable two-stage initialization. Based on above improvements, the proposed Salience DETR achieves significant improvements of +4.0% AP, +0.2% AP, +4.4% AP on three challenging task-specific detection datasets, as well as 49.2% AP on COCO 2017 with less FLOPs. The code is available at https://***/xiuqhou/Salience-DETR.

关键词： Computer vision Filtering Redundancy Semantics Object detection Detectors Transformers

来源：评论

学校读者我要写书评

暂无评论

A Dataset and Benchmark for Shape Completion of Fruits for Agricultural robotics

arXiv

引用

arXiv 2024年

作者： Magistri, Federico Läbe, Thomas Marks, Elias Nagulavancha, Sumanth Pan, Yue Smitt, Claus Klingbeil, Lasse Halstead, Michael Kuhlmann, Heiner McCool, Chris Behley, Jens Stachniss, Cyrill The Center for Robotics The University of Bonn Germany The Lamarr Institute for Machine Learning and Artificial Intelligence Germany The Department of Engineering Science The University of Oxford United Kingdom

As the world population is expected to reach 10 billion by 2050, our agricultural production system needs to double its productivity despite a decline of human workforce in the agricultural sector. Autonomous robotic systems are one promising pathway to increase productivity by taking over labor-intensive manual tasks like fruit picking. To be effective, such systems need to monitor and interact with plants and fruits precisely, which is challenging due to the cluttered nature of agricultural environments causing, for example, strong occlusions. Thus, being able to estimate the complete 3D shapes of objects in presence of occlusions is crucial for automating operations such as fruit harvesting. In this paper, we propose the first publicly available 3D shape completion dataset for agricultural vision systems. We provide an RGB-D dataset for estimating the 3D shape of fruits. Specifically, our dataset contains RGB-D frames of single sweet peppers in lab conditions but also in a commercial greenhouse. For each fruit, we additionally collected high-precision point clouds that we use as ground truth. For acquiring the ground truth shape, we developed a measuring process that allows us to record data of real sweet pepper plants, both in the lab and in the greenhouse with high precision, and determine the shape of the sensed fruits. We release our dataset, consisting of almost 7,000 RGB-D frames belonging to more than 100 different fruits. We provide segmented RGB-D frames, with camera intrinsics to easily obtain colored point clouds, together with the corresponding high-precision, occlusion-free point clouds obtained with a high-precision laser scanner. We additionally enable evaluation of shape completion approaches on a hidden test set through a public challenge on a benchmark server. Copyright © 2024, The Authors. All rights reserved.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Towards MR-Safe Concentric Bellows-Based Hydrostatic Linear Actuator for a Needle Driver

Towards MR-Safe Concentric Bellows-Based Hydrostatic Linear ...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Kwan Kit Lin Yufu Qiu Kim Yan Qingpeng Ding Shing Shin Cheng Department of Mechanical and Automation Engineering and CUHK T Stone Robotics Institute The Chinese University of Hong Kong Hong Kong Shun Hing Institute of Advanced Engineering and Multi-Scale Medical Robotics Center The Chinese University of Hong Kong Hong Kong

Magnetic resonance imaging (MRI) is increasingly used for robotic needle-based clinical diagnosis and therapy due to its high resolution and high soft tissue contrast. A needle driver therefore becomes an essential part of these MRI -guided robotic systems to perform in-bore needle placement under continuous MR imaging. However, existing actuator designs for needle drivers are constrained by the high magnetic field and the bore size of the MR scanner, and most lack high- performance capability, especially in terms of large output force and long stroke. In this work, we introduce an MR-safe concentric bellows linear actuator (CBLA) with improved performance over the existing designs. This soft actuator achieves 45N output force, 77 % stroke-length ratio for the bellows, and around 250% stroke-diameter ratio. A mathematical model was built to characterize the behavior of the actuator to provide guidance for the actuator design. The fabrication and experimental evaluation of the actuator are also presented. The results demonstrate the effectiveness of the CBLA design and the strong potential of its application in MR-guided surgical procedures.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Human Hand Gesture Recognition System Using Body Sensor Network 4

Human Hand Gesture Recognition System Using Body Sensor Netw...

引用

4th International Conference on robotics and Automation in Industry, ICRAI 2021

作者： Maheen, Asra Bin Shahzad, Moosa Asif, M. Usama Ahmad, Syed Fasih Zafar, Saqib Maqbool, H. Farhan Usman, Muhammad Arshad, Hira Department of Mechanical Mechatronics And Manufacturing Engineering Faisalabad Pakistan Human-Centered Robotics Lab National Center of Robotics And Automation Lahore Pakistan Pakistan Engineering Council Islamabad Pakistan Department of Mechanical Mechatronics And Manufacturing Engineering Faisalabad Pakistan

ISBN: (纸本)9781665423434

Wearable technology has gained massive popularity for applications in healthcare in recent years. Body sensor network (BSN) includes wearable sensors such as Myoware muscle sensor that can capture electromyography (EMG) signal from human muscles during a movement. This study involves the control of a 3d prosthetic hand using EMG data of basic hand gestures from healthy subjects and operating the prosthetic hand with this classified data. Artificial Neural Networks (ANN) are biologically inspired computational networks which help to build predictive models from large databases. Backpropagation combined with feedforward ANN is used to classify data here. Classification results showed an accuracy of about 91% for four-hand gestures - hand closed, hand open, pointing index finger, and pinch motion. The BSN-based prosthetic hand can facilitate rehabilitation of amputees, perform surgery using recorded EMG data from a surgeon's hands and reduce industrial manpower in the future. © 2021 IEEE.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Swing-Up of a Weakly Actuated Double Pendulum via Nonlinear Normal Modes

Swing-Up of a Weakly Actuated Double Pendulum via Nonlinear ...

引用

European Control Conference (ECC)

作者： Arne Sachtler Davide Calzolari Maximilian Raff Annika Schmidt Yannik P. Wotte Cosimo Della Santina C. David Remy Alin Albu-Schäffer German Aerospace Center (DLR) Institute of Robotics and Mechatronics Münchener Straβe 20 Oberpfaffenhofen Germany Department of Computer Engineering Technical University of Munich (TUM) Garching Germany University of Stuttgart Institute for Nonlinear Mechanics Stuttgart Germany Robotics & Mechatronics Group University of Twente Enschede The Netherlands Department of Cognitive Robotics Delft University of Technology (TU Delft) Delft The Netherlands

ISBN: (数字)9783907144107

ISBN: (纸本)9798331540920

We identify the nonlinear normal modes spawning from the stable equilibrium of a double pendulum under gravity, and we establish their connection to homoclinic orbits through the unstable upright position as energy increases. This result is exploited to devise an efficient swing-up strategy for a double pendulum with weak, saturating actuators. Our approach involves stabilizing the system onto periodic orbits associated with the nonlinear modes while gradually injecting energy. Since these modes are autonomous system evolutions, the required control effort for stabilization is minimal. Even with actuator limitations of less than 1% of the maximum gravitational torque, the proposed method accomplishes the swing-up of the double pendulum by allowing sufficient time.

关键词： Actuators Torque Tensors Modal analysis Motors Orbits Generators

来源：评论

学校读者我要写书评

暂无评论

Dual Multilayer Extension Actuator for High Extension and High Compression Force Applications

Dual Multilayer Extension Actuator for High Extension and Hi...

引用

IEEE International Conference on Soft robotics (RoboSoft)

作者： Jonathan William Ambrose Dylan Sin You Cheah Raye Chen-Hua Yeow Department of Biomedical Engineering Advanced Robotics Center at the National University of Singapore (NUS) Singapore Department of Biomedical Engineering Nanyang Polytechnic (NYP) Singapore

Soft robotic actuators possess several unique characteristics, such as being generally compliant and lightweight, making them suitable for safe human interaction to be deployed in various industries. However, they are often designed for an intended purpose, making them impractical for a different task. This paper aims to introduce a new way of utilising a new class of soft silicone actuators capable of incorporating them into applications that require both pushing and pulling. The dual multilayer extension actuator (DMEA) is compact and lightweight at 8 g, capable of an extension ratio of 300% and a high pulling force-to-weight ratio of 200 with the assistance of vacuum pressure at 30% extension ratio. The fabrication of the DMEA is described in detail and followed by characterisation tests of the DMEA compared to the FEA model. Finally, we showcase the DMEA pulling a 3D-printed arm as a potential assistive device and demonstrate the DMEA versatility by deploying the DMEA onto everyday household products like scissors and kitchen tongs.

关键词： Industries Fabrication Actuators Force Dynamics Soft robotics Nonhomogeneous media

来源：评论

学校读者我要写书评

暂无评论

Automated Leaf-Level Inspection of Crops Combining UAV and UGV Robots 16th

Automated Leaf-Level Inspection of Crops Combining UAV and...

引用

16th European robotics Forum, ERF 2025

作者： Esser, Felix Marks, Elias Magistri, Federico Weyler, Jan Bultmann, Simon Zaenker, Tobias Ahmadi, Alireza Schreiber, Michael Kuhlmann, Heiner McCool, Chris Popović, Marija Stachniss, Cyrill Behnke, Sven Bennewitz, Maren Klingbeil, Lasse Institute of Geodesy and Geoinformation University of Bonn Bonn Germany Autonomous Intelligent Systems University of Bonn Bonn Germany Humanoid Robots Lab University of Bonn Bonn Germany Agricultural Robotics and Engineering University of Bonn Bonn Germany Center for Robotics University of Bonn Bonn Germany

ISBN: (纸本)9783031894701

Using areal and ground-based robotics systems is one possible component on the way to more sustainable crop production. In this work, we show, how a combination of a UAV and ground robot can be used to automate the process of single plant and even leaf-level inspection of crops in the field, as the farmer usually does manually by physically going to the plants. The work integrates UAV-based plant and weed mapping, UGV path planning and execution, viewpoint planning and control of a robot-mounted arm with a multi-camera head, as well as high-resolution 3D reconstruction. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Crops

来源：评论

学校读者我要写书评

暂无评论

Intelligent Recognition Using Ultralight Multifunctional Nano‑Layered Carbon Aerogel Sensors with Human‑Like Tactile Perception

引用

Nano-Micro Letters 2024年第1期16卷 172-186页

作者： Huiqi Zhao Yizheng Zhang Lei Han Weiqi Qian Jiabin Wang Heting Wu Jingchen Li Yuan Dai Zhengyou Zhang Chris RBowen Ya Yang CAS Center for Excellence in Nanoscience Beijing Key Laboratory of Micro‑Nano Energy and SensorBeijing Institute of Nanoenergy and NanosystemsChinese Academy of SciencesBeijing 101400People’s Republic of China Tencent Robotics X Shenzhen 518054People’s Republic of China School of Nanoscience and Technology University of Chinese Academy of SciencesBeijing 100049People’s Republic of China Department of Mechanical Engineering University of BathBath BA27AKUK Center on Nanoenergy Research School of Physical Science and TechnologyGuangxi UniversityNanning 530004People’s Republic of China

Humans can perceive our complex world through multi-sensory *** limited visual conditions,people can sense a variety of tactile signals to identify objects accurately and ***,replicating this unique capability in robots remains a significant ***,we present a new form of ultralight multifunctional tactile nano-layered carbon aerogel sensor that provides pressure,temperature,material recognition and 3D location capabilities,which is combined with multimodal supervised learning algorithms for object *** sensor exhibits human-like pressure(0.04–100 kPa)and temperature(21.5–66.2℃)detection,millisecond response times(11 ms),a pressure sensitivity of 92.22 kPa^(−1)and triboelectric durability of over 6000 *** devised algorithm has universality and can accommodate a range of application *** tactile system can identify common foods in a kitchen scene with 94.63%accuracy and explore the topographic and geomorphic features of a Mars scene with 100%*** sensing approach empowers robots with versatile tactile perception to advance future society toward heightened sensing,recognition and intelligence.

关键词： Multifunctional sensor Tactile perception Multimodal machine learning algorithms Universal tactile system Intelligent object recognition

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：