检索结果-内蒙古大学图书馆

2018 IEEE Transportation and Electrification Conference and Expo, ITEC 2018

作者： Huynh, Phuoc Sang Vincent, Deepa Azeez, Najath Abdul Patnaik, Lalit Williamson, Sheldon S. Dept. of Electrical Computer and Software Engineering Faculty of Engineering and Applied Science University of Ontario Institute of Technology Faculty of Engineering and APplid Science University of Ontario Instiute of Technology ACE-2025 2000 Smicoe Street North OshawaONL1H 7K4 Canada

ISBN: (纸本)9781538630488

A key requirement in inductive power transfer (IPT) systems is primary high-frequency voltage generation. Until recently, two power conversion stages (AC-DC-AC) were required to generate high-frequency voltage in the IPT systems. These systems are usually costly and cumbersome. Matrix AC- AC converters, highlighted by the absence of bulky DC link storage elements, are considered as a potential alternative. The removal of one power conversion stage enhances the system performance in term of efficiency, reliability, size, weight, and cost. Now, AC-AC buck, half-bridge, and full-bridge converters are gaining popularity in IPT applications. However, highly accurate analysis of their performance in the IPT systems is a challenge. In this paper, a simple and accurate mathematical analysis for the AC-AC buck converter supplying a series-series compensated IPT system is given. Performance indicators used for analysis are input power factor and power transfer capability. The accuracy of the analysis is validated through simulation. The analytical results presented in this paper can also be employed to analyze the series-series IPT system fed from other AC-AC matrix converters. © 2018 IEEE.

关键词： Matrix converters

来源：评论

学校读者我要写书评

暂无评论

CFEA: Collaborative feature ensembling adaptation for domain adaptation in unsupervised optic disc and cup segmentation

arXiv

引用

arXiv 2019年

作者： Liu, Peng Kong, Bin Li, Zhongyu Zhang, Shaoting Fang, Ruogu J. Crayton Pruitt Family Dept. of Biomedical Engineering University of Florida GainesvilleFL United States Department of Computer Science UNC Charlotte CharlotteNC United States School of Software Engineering Xi'an Jiaotong University Xi'an China Sensetime Research

Recently, deep neural networks have demonstrated comparable and even better performance with board-certified ophthalmologists in well-annotated datasets. However, the diversity of retinal imaging devices poses a significant challenge: domain shift, which leads to performance degradation when applying the deep learning models to new testing domains. In this paper, we propose a novel unsupervised domain adaptation framework, called Collaborative Feature Ensembling Adaptation (CFEA), to effectively overcome this challenge. Our proposed CFEA is an interactive paradigm which presents an exquisite of collaborative adaptation through both adversarial learning and ensembling weights. In particular, we simultaneously achieve domain-invariance and maintain an exponential moving average of the historical predictions, which achieves a better prediction for the unlabeled data, via ensembling weights during training. Without annotating any sample from the target domain, multiple adversarial losses in encoder and decoder layers guide the extraction of domain-invariant features to confuse the domain classifier and meanwhile benefit the ensembling of smoothing weights. Comprehensive experimental results demonstrate that our CFEA model can overcome performance degradation and outperform the state-of-the-art methods in segmenting retinal optic disc and cup from fundus images. Code is available at https://***/cswin/AWC. Copyright © 2019, The Authors. All rights reserved.

关键词： Ophthalmology

来源：评论

学校读者我要写书评

暂无评论

LOCALIZED PATTERNS OF THE CUBIC-QUINTIC SWIFT-HOHENBERG EQUATIONS WITH TWO SYMMETRY-BREAKING TERMS

引用

Annals of Applied Mathematics 2018年第1期34卷 94-110页

作者： Yancong Xu Tianzhu Lan Zhenxue Wei Dept.of Math. Hangzhou Normal University School of Computer Science and Software Engineering East China Normal University

Homoclinic snake always refers to the branches of homoclinic orbits near a heteroclinic cycle connecting a hyperbolic or non-hyperbolic equilibrium and a periodic orbit in a reversible variational system. In this paper, the normal form of a Swift-Hohenberg equation with two different symmetry-breaking terms （non-reversible term and non-k-symmetry term） are investigated by using multiple scale method, and their bifurcation diagrams are initially studied by numerical simulations. Typically, we predict numerically the existence of so- called round-snakes and round-isolas upon particular two symmetric-breaking perturbations.

关键词： round-snakes round-isolas normal form Swift-Hohenberg equation localized patterns

来源：评论

学校读者我要写书评

暂无评论

Public policy challenges: An RE perspective 7

Public policy challenges: An RE perspective

引用

7th International Workshop on Requirements engineering for Sustainable Systems, RE4SuSy 2018

作者： Callele, David Penzenstadler, Birgit Wnuk, Krzysztof Dept. of Computer Science University of Saskatchewan Canada Dept. of Comp. Eng. and Comp. Sci. CSULB Long Beach United States Department of Software Engineering Blekinge Institute of Technology Karlskrona Sweden

In this perspective paper, we investigate the parallels between public policy and IT projects from the perspective of traditional RE practice. Using the mainstream media as an information source (as would an average citizen), over a period of approximately one year we captured documents that presented analyses of public policy issues. The documents were categorized into eight topic areas, then analyzed to identify patterns that RE practitioners would recognize. We found evidence of policy failures that parallel project failures traceable to requirements engineering problems. Our analysis revealed evidence of bias across all stakeholder groups, similar to the rise of the "beliefs over facts" phenomenon often associated with "fake news". We also found substantial evidence of unintended consequences due to inadequate problem scoping, terminology definition, domain knowledge, and stakeholder identification and engagement. Further, ideological motivations were found to affect constraint definitions resulting in solution spaces that may approach locally optimal but may not be globally optimal. Public policy addresses societal issues;our analysis supports our conclusion that RE techniques could be utilized to support policy creation and implementation. © 2018 SPIE. All rights reserved.

关键词： Public policy

来源：评论

学校读者我要写书评

暂无评论

Android Malware Detection Using Feature Selections and Random Forest

Android Malware Detection Using Feature Selections and Rando...

引用

International Conference on software Security and Assurance (ICSSA)

作者： Taehoon Eom Heesu Kim SeongMo An Jong Sou Park Dong Seong Kim Dept. of Computer Engineering Korea Aerospace University Seoul South Korea Dept. of Computer Science and Software Engineering The University of Queensland Australia

ISBN: (数字)9781538692103

ISBN: (纸本)9781538692110

Malicious software (Malware) applications in Android ecosystem is one of the critical issues. Manual detection of malware is not cost-effective and cannot keep up with the fast evolution of malware development in Android. A machine learning based malware detection has attempted to automate the detection of malware in Android. In this paper, we present new Android malware detection methods. The main idea of our proposed approach is to use three different feature selection methods before malware detection model using a machine learning algorithm is constructed. We used both Malware Genome Project dataset and our own crawled dataset to show the effectiveness of the proposed methods.

关键词： Malware Feature extraction Forestry Androids Humanoid robots Machine learning algorithms Smart phones

来源：评论

学校读者我要写书评

暂无评论

Directional Motion-based Interfaces for Virtual and Augmented Reality Head-mounted Displays

Directional Motion-based Interfaces for Virtual and Augmente...

引用

International computers, Signals and Systems Conference (ICOMSSC)

作者： Wenge Xu Haining Liang Yong Yue Dept. of Computer Science and Software Engineering Xi’an Jiaotong Liverpool University Suzhou China

ISBN: (纸本)9781538667521

This research, conducts an experiment to investigate whether there is a difference in user performance and preference for two types of head-mounted displays (HMDs) when users need to perform directional motion movements such as moving one foot forward and backward (or leftward or rightward). The two types of HMDs we have considered are virtual reality (VR) and augmented reality (AR), which represent the two most commonly marketed HMDs. The AR device chosen for this research is the Meta 2 while the chosen VR device is the Oculus RIFT CV1. The results of our experiment show that there is a close significant difference on task completion time between AR and VR condition when users perform motion direction tasks. Also, no significant effect has been found on the accuracy of making these motion movements between the AR and VR conditions. In terms of user preference, the results show that there is no significant effect on workload, motion sickness, immersion, and user experience. These results suggest that both AR and VR HMDs are suitable for interfaces that can rely on body motions like tapping on the floor using one foot in any of the directions around the user.

关键词： Task analysis Analysis of variance User experience Performance evaluation Atmospheric measurements Particle measurements Time measurement

来源：评论

学校读者我要写书评

暂无评论

Improving Generative Adversarial Networks with Adaptive Control Learning

Improving Generative Adversarial Networks with Adaptive Cont...

引用

IEEE Visual Communications and Image Processing (VCIP)

作者： Xiaohan Ma Rize Jin Kyung-Ah Sohn JoonYoung Paik Jing Sun Tae-Sun Chung Dept. of Software Ajou University Suwon Korea School of Computer Science and Software Engineering Tianjin Polytechnic University Tianjin China

ISBN: (纸本)9781538644591;9781538644584

Generative adversarial networks (GANs) are well known both for being unstable to train and for the problem of mode collapse, particularly when trained on data collections containing a diverse set of visual objects. This study introduces an adaptive hyper-parameter learning procedure for GANs as an alternative to the existing static approach. The proposed procedure is designed to mitigate the impact of instability and saturation in the original by dynamically adjusting the ratio of the training steps of both the generator and discriminator. To accomplish this, we track and analyze stable training curves of relatively narrow datasets and use them as the target fitting lines when training more diverse data collections. Experimental results show that the proposed model improves the stability and generates more realistic images.

关键词： Training Gallium nitride Generators Generative adversarial networks Adaptive control Adaptation models Stability analysis

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Context Based Classification of Twitter 4

Comparative Analysis of Context Based Classification of Twit...

引用

4th International Conference on Advances in Computing, Communication and Automation, ICACCA 2018

作者： Tiwana, Muhammad Salman Javeed, Farooq Lali, Muhammad Ikramullah Dar, Hafsa Bilal, Muhammad Dept. of CS and IT University of Sargodha Sargodha Pakistan Dept. of Computer Science University of Gujrat Gujrat Pakistan Dept. of Software Engineering University of Gujrat Gujrat Pakistan School of Computing and IT Taylor's University Subang Jaya Malaysia Centre for Data Science and Analytics Taylor's University Subang Jaya Malaysia

ISBN: (纸本)9781538671672

Social media offers online data source where internet users' share their opinions, views and discuss any news or events that happen around the world. Twitter is a popular social media which effectively takes part in any mega event and is used before, during and after live events. Some tweets become trend based on their popularity. The trending tweets are classified by twitter into categories like: Sports (S), Entertainment (E), Politics (P), and Technology (T) which are used in our work. We have used context based meanings to further improve the trending tweet classification results. In this article, we have verified the twitter's automated classification results with manual classification by experts. Furthermore, we compared the classification results with SVM, Linear Regression, Logistic Regression and Naive Bayes implemented in Weka tool. We found that twitter's classifier works with almost 89% accuracy in our dataset. This work would contribute in improving misclassification of data of different categories on social media. © 2018 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Underrepresented Groups Threats to Belonging in Computing

Underrepresented Groups Threats to Belonging in Computing

引用

Research in Equity and Sustained Participation in engineering, Computing, and Technology (RESPECT)

作者： Marlon Mejias Ketly Jean-Pierre Gloria Washington Legand Burge Dept. of Software & Information Systems University. of North Carolina at Charlotte Charlotte NC USA Post-Doctoral Fellow - Process Data Educcational Testing Services Princeton NJ USA Dept. of Electrical Engineering & Computer Science Howard Unversity Washingington D.C. USA

Increasing the number of students exposed to computer science (CS) may not increase throughput in the pipeline if barriers to retention are not addressed. One major barrier is student's perception of belonging. To begin creating theoretical grounds for a common discourse to successfully matriculating underrepresented students, and retaining them in the field, threats to their belonging (barriers to retention) and how a survey of literature suggest they impact CS are presented.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deep learning for automated image classification of seismic damage to built infrastructure 11

Deep learning for automated image classification of seismic ...

引用

11th National Conference on Earthquake engineering 2018: Integrating science, engineering, and Policy, NCEE 2018

作者： Patterson, B. Leone, G. Pantoja, M. Behrouzi, A. Dept. of Computer Science & Software Engineering California Polytechnic State University - San Luis Obispo San Luis ObispoCA93407 United States Dept. of Architectural Engineering California Polytechnic State University - San Luis Obispo San Luis ObispoCA93407 United States

ISBN: (纸本)9781510873254

The amount of structural damage image data produced in the aftermath of an earthquake can be staggering. It is challenging for a few human volunteers to efficiently filter and tag these images with meaningful damage information. The proposed solution is to automate post-earthquake reconnaissance image tagging activities by training a computer algorithm to classify each occurrence of damage per building material and structural member type. The approach is based on deep learning (DL), a subset of machine learning loosely based on the operation of a biologic neural system, which aims to learn and extract accurate representations from large data sets. DL algorithms are data driven;improving with increased training data. Thanks to the vast amount of data available and advances in computer architectures, DL has become one of the most popular image classification algorithms producing results comparable to and in some cases superior to human experts. The authors implemented a DL algorithm to automatically identify multiple damage types and associated structural members in a single image by adapting a pre-trained deep residual network. The algorithm was tested as follows: (i) binning building images as damage-no damage (88% accuracy), (ii) drawing a bounding box around damage in buildings (85% accuracy) and short/captive reinforced concrete columns with shear damage (77% accuracy). The lower accuracy of correctly identifying a target region in an image (test ii) compared to simple binning (test i) is anticipated since it is a more complex problem and there is a more limited number of expertly tagged training images (200 count) for shear damage-short column condition being studied. The research team expects algorithm accuracy will improve as with training on additional images tagged for certain damage-structure pairs by a diverse set of experts. Copyright © 2018 Earthquake engineering Research Institute. All rights reserved.

关键词： Image enhancement

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：