检索结果-内蒙古大学图书馆

2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020

作者： Hu, Rui Cai, Jiaming Zheng, Wangjie Yang, Yang Shen, Hong-Bin Shanghai Jiao Tong University Key Lab. of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Department of Computer Science and Engineering Shanghai200240 China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China Shanghai Jiao Tong University Department of Bioinformatics and Biostatistics Shanghai200240 China

ISBN: (纸本)9781728162157

Cryo-electron microscopy (cryo-EM) has become a mainstream technology for solving spatial structures of biomacromolecules, while the processing of cryo-EM images is a very challenging task. One of the great challenges is the high noise in the images. A common method is to cluster the images with close projecting angles to get mean images, which are used for 3D reconstruction. However, due to the extremely low signal-to-noise-ratio, common clustering methods often fail to obtain high-quality mean images, leading to poorly reconstructed structures. In this study, we present a new unsupervised learning framework, called NiuEM, to discriminate images captured from different angles and yield cluster-mean images. NiuEM first generates pseudo-labels and then exploits both contrastive loss and cross-entropy loss for training convolutional layers to learn feature representations. Moreover, the pseudo-labels are updated iteratively to enhance the reliability of labels. We assess the performance of NiuEM on four data sets via both visualized and quantitative experiments. Especially, two kinds of metrics are adopted to measure the performance, regarding the clustering quality and the resolution of reconstructed 3D models, respectively. The experimental results show that NiuEM achieves very competitive clustering accuracy in the comparison with the state-of-the-art image clustering methods. Moreover, the cluster mean images yielded by NiuEM lead to better initial 3D models compared with the mainstream reconstruction tools. © 2020 IEEE.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Simultaneously Transmitting and Reflecting Reconfigurable Intelligent Surface Assisted NOMA Networks

arXiv

引用

arXiv 2021年

作者： Yue, Xinwei Xie, Jin Liu, Yuanwei Han, Zhihao Liu, Rongke Ding, Zhiguo The Key Laboratory of Information and Communication Systems Ministry of Information Industry China The Key Laboratory of Modern Measurement & Control Technology Ministry of Education Beijing Information Science and Technology University Beijing100101 China The School of Electronic Engineering and Computer Science Queen Mary University of London LondonE1 4NS United Kingdom The School of Electronic and Information Engineering Beihang University Beijing100191 China The Department of Electrical Engineering Princeton University Princeton United States The School of Electrical and Electronic Engineering The University of Manchester Manchester United Kingdom

Simultaneously transmitting/refracting and reflecting reconfigurable intelligent surface (STAR-RIS) has been introduced to achieve full coverage area. This paper investigate the performance of STAR-RIS assisted non-orthogonal multiple access (NOMA) networks over Rician fading channels, where the incidence signals sent by base station are reflected and transmitted to the nearby user and distant user, respectively. To evaluate the performance of STAR-RIS-NOMA networks, we derive new approximate expressions of outage probability and ergodic rate for a pair of users, in which the imperfect successive interference cancellation (ipSIC) and perfect SIC (pSIC) schemes are taken into consideration. Based on the asymptotic expressions, the diversity orders of the nearby user with ipSIC/pSIC and distant user are achieved carefully. The high signal-to-noise ratio slopes of ergodic rates for nearby user with pSIC and distant user are equal to one and zero, respectively. In addition, the system throughput of STAR-RIS-NOMA is discussed in delay-limited and delay-tolerant modes. Simulation results are provided to verify the accuracy of the theoretical analyses and demonstrate that: 1) The outage probability of STAR-RIS-NOMA outperforms that of STAR-RIS assisted orthogonal multiple access (OMA) and conventional cooperative communication systems;2) With the increasing of reflecting elements K and Rician factor κ, the STAR-RIS-NOMA networks are capable of attaining the enhanced performance;and 3) The ergodic rates of STAR-RIS-NOMA are superior to that of STAR-RIS-OMA. Copyright © 2021, The Authors. All rights reserved.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

An Improved Phasor Domain Parameter-Free Fault Location Algorithm on Untransposed Lines

An Improved Phasor Domain Parameter-Free Fault Location Algo...

引用

IEEE PES Transmission and Distribution Conference and Exposition

作者： Dian Lu Yu Liu Boqi Xie Rui Fan Liangyi Sun School of Information Science and Technology ShanghaiTech University Shanghai China Key Laboratory of Control of Power Transmission and Conversion Ministry of Education Shanghai China School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta GA U.S School of Engineering and Computer Science University of Denver Denver CO U.S

ISBN: (数字)9781728145471

ISBN: (纸本)9781728145488

This paper proposes a parameter-free fault location algorithm on untransposed transmission lines. Synchrophasor voltage and current measurements at both terminals of the line before and during faults are typically required. The method first builds the transmission line models before and during the fault, which describe all the physical laws that the line should obey. The fault location as well as transmission line parameters are introduced as unknown variables in the models. Afterwards, the fault location is obtained by accurately solving the unknown variables in the models. Compared to the existing parameter-free fault location methods on untransposed lines, the proposed method fully considers the lumped shunt capacitance matrix and directly solves the fault location without further assumptions. Numerical experiments in PSCAD/EMTDC proved that the proposed algorithm is more accurate than the existing untransposed parameter-free fault location method, independent of fault locations, fault types and fault impedances.

关键词： Fault location Power transmission lines Transmission line measurements Transmission line matrix methods Admittance Current measurement Capacitance

来源：评论

学校读者我要写书评

暂无评论

Differentiator-based disturbance observer 8

Differentiator-based disturbance observer

引用

8th IEEE Data Driven control and Learning Systems Conference, DDCLS 2019

作者： Li, Xiangyang Gao, Zhiqiang Ai, Wei Tian, Senping School of Automation Science and Engineering South China University of Technology Guangzhou510641 China Center for Advanced Control Technologies Cleveland State University Department of Electrical and Computer Engineering ClevelandOH44115 United States

ISBN: (纸本)9781728114545

The analysis of nonlinear disturbance observer (NDO) is first given in detail and an equivalent differentiator structure is used to explain the NDO. The design procedure of NDO is given according to this equivalent structure. Then the structure and principle of the generalized differentiator-based disturbance observer (DBDO) is given and compared with NDO in their structure and algorithm. As one of DBDO, tracking differentiator (TD) based disturbance observer is applied to compare NDO in simulations. Simulations for disturbance estimation show that both NDO and TD based disturbance observer have very effective disturbance estimation performance but the later has less design restriction. © 2019 IEEE.

关键词： Nonlinear disturbance observer (NDO) Differentiator-based disturbance observer (DBDO) Tracking differentiator (TD) Disturbance observer-based control (DOBC) NDO disturbance observer design process DIFFERENTIATORS

来源：评论

学校读者我要写书评

暂无评论

Automated Injuries Classification on Postmortem Images: A Deep Learning Approach with Different Neural Networks Architectures

SSRN

引用

SSRN 2023年

作者： Kalinowska, Kamila Barbara Zawieska, Dorota Puchala, Sebastian Markiewicz, Jakub Stefan Kot, Patryk Wilkowski, Artur Krajewski, Pawel Fudalej, Marcin Soltyszewski, Ireneusz Department of Photogrammetry Remote Sensing and Spatial Information Systems Warsaw University of Technology Politechniki 1 Square Warsaw00-661 Poland Institute of Computer Science Warsaw University of Technology Politechniki 1 Square Warsaw00-661 Poland Research Institute Liverpool John Moores University Byrom Street LiverpoolL3 3AF United Kingdom Institute of Control and Computation Engineering Warsaw University of Technology Politechniki 1 Square Warsaw00-661 Poland Forensic Medicine Department Medical University of Warsaw 61 Żwirki and Wigury Street Warsaw02-091 Poland

The paper discusses a deep learning approach for automatically detecting the two most important classes of injuries (bruises and abrasions) with the use of semantic segmentation. First, a dataset of postmortem images of injuries is completed and fitting data pre-processing, and annotation methods for algorithm training and evaluation are developed. Second, end-to-end U-Net, FPN and LinkNet semantic segmentation network-based injuries detection algorithms are implemented on EfficientNetB3 and ResNet50 backbones. Third, an optimisation strategy for the network is employed to improve the algorithm's detection performance based on the data features of injuries. This technique entails data augmentation utilising a custom loss function and a combination of image transformation and class balancing. The experimental results demonstrated promising performance with high sensitivity and specificity for our best model called TIPPI (92,7% and 98,9%, respectively). TIPPI may be used to automatically analyse postmortem images. © 2023, The Authors. All rights reserved.

关键词： Metadata

来源：评论

学校读者我要写书评

暂无评论

Lightweight Distributed Attribute Based Keyword Search System for Internet of Things 12th

Lightweight Distributed Attribute Based Keyword Search Syste...

引用

12th International Conference on Security, Privacy, and Anonymity in Computation, Communication, and Storage, SpaCCS 2019

作者： Long, Jiahuan Zhang, Ke Wang, Xiaofen Dai, Hong-Ning School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu611731 China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China Science and Technology on Electronic Information Control Laboratory Chengdu610000 China Department of Information Technology Macau University of Science and Technology Macau China

ISBN: (纸本)9783030248994

Internet of Things (IoT) is a promising networking paradigm that connects various kinds of sensors and exchanges data from smart devices. Since IoT always related to user’s daily life, the problems of security and privacy are of great importance. Presently, the attribute based encryption (ABE) is a popular solution to guarantee the fine-grained sharing of encrypted data in IoT. In this paper, an attribute based keyword search with lightweight decryption in multi-authority (ABKS-LD-MA) is proposed. Our system supports multi-keyword search in cloud by using searchable attribute based encryption. We also integrate the lightweight decryption to searchable ABE scheme that largely reduces the computing overhead for users. Furthermore, our ABKS-LD-MA scheme supports multi-authority scenario, which is more adaptive to the real IoT environment. The experiment analysis shows that our scheme has relatively lowered the communication cost on IoT devices. © 2019, Springer Nature Switzerland AG.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

3D Hierarchical Refinement and Augmentation for Unsupervised Learning of Depth and Pose from Monocular Video

arXiv

引用

arXiv 2021年

作者： Wang, Guangming Zhong, Jiquan Zhao, Shijie Wu, Wenhua Liu, Zhe Wang, Hesheng Department of Automation Institute of Medical Robotics Key Laboratory of System Control and Information Processing Ministry of Education Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education Shanghai Engineering Research Center of Intelligent Control and Management Shanghai Jiao Tong University Shanghai200240 China Department of Engineering Mechanics Shanghai Jiao Tong University Shanghai200240 China Department of Computer Science and Technology University of Cambridge CambridgeCB2 1TN United Kingdom

Depth and ego-motion estimations are essential for the localization and navigation of autonomous robots and autonomous driving. Recent studies make it possible to learn the per-pixel depth and ego-motion from the unlabeled monocular video. A novel unsupervised training framework is proposed with 3D hierarchical refinement and augmentation using explicit 3D geometry. In this framework, the depth and pose estimations are hierarchically and mutually coupled to refine the estimated pose layer by layer. The intermediate view image is proposed and synthesized by warping the pixels in an image with the estimated depth and coarse pose. Then, the residual pose transformation can be estimated from the new view image and the image of the adjacent frame to refine the coarse pose. The iterative refinement is implemented in a differentiable manner in this paper, making the whole framework optimized uniformly. Meanwhile, a new image augmentation method is proposed for the pose estimation by synthesizing a new view image, which creatively augments the pose in 3D space but gets a new augmented 2D image. The experiments on KITTI demonstrate that our depth estimation achieves state-of-the-art performance and even surpasses recent approaches that utilize other auxiliary tasks. Our visual odometry outperforms all recent unsupervised monocular learning-based methods and achieves competitive performance to the geometry-based method, ORB-SLAM2 with back-end optimization. Copyright © 2021, The Authors. All rights reserved.

关键词： Robots

来源：评论

学校读者我要写书评

暂无评论

A lightweight perception module for planning purposes

A lightweight perception module for planning purposes

引用

International Conference on Emerging Technologies and Factory Automation (ETFA)

作者： Muhayy Ud Din Jan Rosell Sohail Bukhari Mansoor Ahmad Wajahat M Qazi Institute of Industrial and Control Engineering (IOC) Universitat Politècnica de Catalunya (UPC) Barcelona Tech Spain Department of Computer Science COMSATS University Islamabad Lahore Campus Pakistan

ISBN: (数字)9781728189567

ISBN: (纸本)9781728189574

Sensing is an essential component for robots to perform the manipulation tasks in real environments. This study proposes a lightweight deep-learning-based sensing modules which allows the robots to automatically model the workspace for manipulation planning. This sensing module is developed as a part of our ongoing manipulation planning framework. It will be used to enhance the sensing accuracy and make it capable of planning the manipulation tasks in real environments. The retrained model is further trained over commonly used objects to enhance the prediction accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Game-based Learning Approach to Cybersecurity

Game-based Learning Approach to Cybersecurity

引用

IEEE Education engineering (EDUCON)

作者： Llanos Tobarra Antonio Pérez Trapero Rafael Pastor Antonio Robles-Gómez Roberto Hernández Andrés Duque Jesús Cano Communication and Control System Department Spanish National Distance University (UNED) Madrid Spain Student Computer Science Engineering School Spanish National Distance University (UNED) Madrid Spain

ISBN: (数字)9781728109305

ISBN: (纸本)9781728109312

The inclusion of practical activities to train critical thinking skills is very relevant in the case of engineering subjects. In particular, it becomes a key element in the context of cybersecurity, as our case is. Therefore, this paper analyses a gambling experience in a cybersecurity subject within a computer engineering degree, in order to assess whether the spaced learning methodology contributes to the improvement of learning. The gamification experience is carried out through a Flag Capture Competition (CTF) among the enrolled students in the course. The use of the platform has also been evaluated in an exploratory way.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-model least squares-based recomputation framework for large data analysis

arXiv

引用

arXiv 2021年

作者： Zhang, Wandong Wu, Q.M. Jonathan Yang, Yimin Zhao, W.G. Will Wang, TianLei Zhang, Hui Department of Electrical and Computer Engineering University of Windsor Windsor Canada Department of Computer Science Lakehead University Thunder Bay Canada Vector Institute for Artificial Intelligence Toronto Canada Faculty of Business Administration Lakehead University Canada Artificial Intelligence Group CEGEP de Ste Foy Canada Institute of Information and Control Hangzhou Dianzi University Hangzhou China School of Robotics Hunan University Changsha China

Most multilayer least squares (LS)-based neural networks, such as deep random vector functional link (RVFL), are structured with two separate stages: unsupervised feature encoding and supervised pattern classification. Once the unsupervised learning is finished, the latent encoding would be fixed without supervised fine-tuning. However, in complex tasks such as handling the ImageNet dataset, there are often many more clues that can be directly encoded, while the unsupervised learning, by definition, cannot know exactly what is useful for a certain task. There is a need to retrain the latent space representations in the supervised classification stage to learn some clues that unsupervised learning has not yet learned. In particular, the error matrix from the output layer is pulled back to each hidden layer, and the parameters of the hidden layer are recalculated with Moore-Penrose (MP) inverse for more generalized representations. In this paper, a recomputation-based multilayer network using MP inverse (RML-MP) is developed. A sparse RML-MP (SRML-MP) model to boost the performance of RML-MP is then proposed. The experimental results with varying training samples (from 3 K to 1.8 M) show that the proposed models provide better generalization performance than most representation learning algorithms. Copyright © 2021, The Authors. All rights reserved.

关键词： Signal encoding

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：