检索结果-内蒙古大学图书馆

computers and Artificial Intelligence Technologies (CAIT), International Conference on

作者： Zhengbin Zhang Zhenhao Xu Xingsheng Gu Juan Xiong School of Information Science and Engineering East China University of Science and Technology Shanghai China School of Optical Electircal and Computer Engineering University of Shanghai for Science and Technology Shanghai China

Semantic segmentation empowers various real-world applications. Nevertheless, the substantial computational cost, such as O(k 2 ) time complexity associated with the number of tokens in multi-headed self-attention, poses challenges for deploying these models on edge devices with constrained hardware resources. This paper introduces a novel family of backbones designed for real-time semantic segmentation, referred to as Linear and Re-parameter Vision Transformer (LARFormer). Particularly, we introduce a Re-parameter Mobile Block (RMB), which employs three branches during training and a single branch during inference. Furthermore, we introduce Linear Separable Self-Attention(LSSA), which reduces the computational complexity from O(k 2 ) to O(k). Extensive experiments on the ADE20K dataset and Pascal VOC 2012 dataset demonstrate the effectiveness of the proposed LARFormer by achieving a promising trade-off between segmentation accuracy and inference speed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Functional and effective connectivity in major depressive disorder patients with suicidal thoughts: A resting-state fMRI study

Functional and effective connectivity in major depressive di...

引用

Iranian Conference of Biomedical engineering (ICBME)

作者： Morteza Fattahi Reza Rostami Gholam-Ali Hossein-Zadeh School of Electrical and Computer Engineering University of Tehran Tehran Iran School of Psychology and Educational Sciences University of Tehran Tehran Iran School of Electircal and Computer Engineering University of Tehran Tehran Iran

ISBN: (纸本)9781665499859

Major Depressive Disorder (MDD) is one of the most prevalent psychological disorders all over the world. Suicide ideation is the most dangerous and crucial consequence of this mental disturbance that can lead to suicide attempting. In this study, we compared both functional and effective brain connectivity of female MDD patients having suicide thoughts with female healthy controls. We used COMBI algorithm for identifying independent components in resting state fMRI data (rsfMRI) of 32 female subjects (16 MDDs with suicide ideation and 16 healthy controls). We performed Group Information Guided Independent Component Analysis (GIG-ICA) as a back-reconstruction step and finally, extracted the static and Dynamic Functional Connectivity (DFC) for all participants. Besides that, Dynamic Effective Connectivity (DEC) was characterized using Dynamic Granger Causality (DGC). Significant connectivity differences were detected between some network pairs including: Auditory Network with right Executive Control Network (ECN), ventral Default Mode Network (DMN) with Visuospatial, right ECN with anterior Salience Network (SN), ventral DMN with Precuneus, dorsal DMN with Precuneus, Language with dorsal DMN, Language with posterior SN and Primary Visual with dorsal DMN. Except the connectivity difference between right ECN and anterior Salience network, other connectivity changes were consistent with previous studies. As far as we know, this is the first study to investigate the effective connectivity by the information extracted from ICA in this specific study group.

关键词： Visualization Heuristic algorithms Psychology Focusing Independent component analysis Functional magnetic resonance imaging Depression

来源：评论

学校读者我要写书评

暂无评论

Minimax Pareto Fairness: A multi objective perspective

arXiv

引用

arXiv 2020年

作者： Martinez, Natalia Bertran, Martin Sapiro, Guillermo Department of Electircal and Computer Engineering Duke University United States

In this work we formulate and formally characterize group fairness as a multi-objective optimization problem, where each sensitive group risk is a separate objective. We propose a fairness criterion where a classifier achieves minimax risk and is Pareto-efficient w.r.t. all groups, avoiding unnecessary harm, and can lead to the best zero-gap model if policy dictates so. We provide a simple optimization algorithm compatible with deep neural networks to satisfy these constraints. Since our method does not require test-time access to sensitive attributes, it can be applied to reduce worst-case classification errors between outcomes in unbalanced classification problems. We test the proposed methodology on real case-studies of predicting income, ICU patient mortality, skin lesions classification, and assessing credit risk, demonstrating how our framework compares favorably to other approaches. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

IoT-based irrigation management for smallholder farmers in rural Sub-Saharan Africa 11

IoT-based irrigation management for smallholder farmers in r...

引用

11th International Conference on Emerging Ubiquitous Systems and Pervasive Networks, EUSPN 2020

作者： Nigussie, Ethiopia Olwal, Thomas Musumba, George Tegegne, Tesfa Lemma, Atli Mekuria, Fisseha Department of Future Technologies Turku Finland Department of Electrical Engineering Tshwane University of Technology Pretoria South Africa School of Computer Science and Information Technology Dedan Kimathi University of Technology Nyeri Kenya Center for ICT for Development Bahir Dar University Bahir Dar Ethiopia School of Electircal Engineering Haramaya University Dire Dawa Ethiopia Council for Scientific and Industrial Research Pretoria South Africa

Ensuring food security has become a challenge in Sub-Saharan Africa (SSA) due to combined effects of climate change, high population growth, and relying on rainfed farming. Governments are establishing shared irrigation infrastructure for smallholder farmers as part of the solutions for food security. However, the irrigated farms often failed to achieve the expected crop yield. This is partly due to lack of water management system in the irrigation infrastructure. In this work, IoT-based irrigation management system is proposed after investigating problems of irrigated farmlands in three SSA countries, Ethiopia, Kenya, and South Africa as case studies. Resource-efficient IoT architecture is developed that monitors soil, microclimate and water parameters and performs appropriate irrigation management. Indigenous farming and expert knowledge, regional weather information, crop and soil specific characteristics are also provided to the system for informed-decision making and efficient operation of the irrigation management system. In SSA, broadband connectivity and cloud services are either unavailable or expensive. To tackle these limitations, data processing, network management, irrigation decisions and communication to the farmers are carried out locally, without the involvement of any back-end servers. Furthermore, the use of green energy sources and resource-aware intelligent data analysis algorithm is studied. It is expected that an intelligent data analysis will help to discover new knowledge that support further development of agricultural expertise. The proposed IoT-based irrigation management system is expected to contribute towards long term and sustainable high crop yield with minimum resource consumption and impact to the biodiversity around the case farmlands. © 2020 The Authors. Published by Elsevier B.V.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Narrowband IoT: An appropriate solution for developing countries

arXiv

引用

arXiv 2019年

作者： Routray, Sudhir K. Hussein, Habib Mohammed Department of Electrical and Computer Engineering College of Electircal and Mechanical Engineering Addis Ababa Science and Technology University Addis Ababa Ethiopia

Internet of things (IoT) is very much attractive for several sensor based applications. It provides large coverage of the services with small amount of resources. Its applications span from the ordinary scenarios such as sensing in the common digital ecosystem to the far more complicated processes of modern manufacturing, agriculture, security provisioning, location tracking and health care. Several types of IoTs have been proposed for the recent applications. Narrowband IoT (NBIoT) is one of the economical versions of the IoTs. It is a low power wide area network technology and thus suitable for resource limited scenarios. In the developing countries, the resources are scarce and economical solutions are always preferable. Therefore, NBIoT is an attractive solution for the developing countries. In this article, we present its features and functions which make it suitable for developing countries. We also provide several sector based analysis which are suitable for the NBIoT deployment. Copyright © 2019, The Authors. All rights reserved.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Satellite Based IoT Networks for Emerging Applications

arXiv

引用

arXiv 2019年

—Internet of things (IoT) has been proven to be a ubiquitous technology for several existing and new applications. It can provide both accuracy and sustainability in the emerging services and applications. IoT has several advantages and it can provide different levels of coverage in different geographical locations and contexts. There are several applications in which the wide coverage, low power and reliability of the communication have very high priorities. Conglomeration of IoT and satellite networks can make it very attractive for these applications. In this article, we present the general features, motivation and deployment issues of satellite based IoT networks. We present the applications where these satellite based IoT networks can play important roles. We also present the hybridization of narrowband IoT and satellite networks which can provide a long term sustainable network solution for several applications. Copyright © 2019, The Authors. All rights reserved.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

IoT-based Irrigation Management for Smallholder Farmers in Rural Sub-Saharan Africa

引用

Procedia computer Science 2020年 177卷 86-93页

作者： Ethiopia Nigussie Thomas Olwal George Musumba Tesfa Tegegne Atli Lemma Fisseha Mekuria Department of Future Technologies Turku Finland Department of Electrical Engineering Tshwane University of Technology Pretoria South Africa School of Computer Science and Information Technology Dedan Kimathi University of Technology Nyeri Kenya Center for ICT for Development Bahir Dar University Bahir Dar Ethiopia School of Electircal Engineering Haramaya University Dire Dawa Ethiopia Council for Scientific and Industrial Research Pretoria South Africa

关键词： Internet of things smart irrigation smallholder farmers intelligent data analysis low-cost sensors

来源：评论

学校读者我要写书评

暂无评论

Prediction and estimation of next demands of cloud users based on their comments in CRM and previous usages

Prediction and estimation of next demands of cloud users bas...

引用

2018 International Conference On Communication, Computing and Internet of Things, IC3IoT 2018

作者： Darbandi, Mehdi Haghgoo, Saeed Hajiali, Mahdi Khabir, Alireza Department of Electrical and Electronic Engineering Eastern Meditteranean University via Mersin 10 Gazimagusa Turkey Department of Computer Engineering Islamic Azad University Islamshahr Branch Iran Department of Computer Engineering and Information Technology Amirkabir University of Technology Tehran Iran Faculty of Electircal Engineering Islamic Azad University Tabriz Iran

ISBN: (纸本)9781538624586

Today, IT technologies and the internet are become an essential factors in human life's. The demands level of the users are growing extensively, they need more generic software and hardware services, also they need, security in their processes and data transfers, dynamic accessibility, swift processing and also saving more money in their works. In this paper, we want to present new method for estimating and predicting the next user demands, and provide those services before direct request of the users. Our algorithm is based on Kalman filtering, and it used background usages of the users along with their suggestions in CRM section to be able to predict their next needs. By providing the next needs of the users, before their request, we can improve the level of customer satisfaction and also earn more money via suggesting the new software and applications. © 2018 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

Prediction and Estimation of Next Demands of Cloud Users based on their Comments in CRM and Previous usages

Prediction and Estimation of Next Demands of Cloud Users bas...

引用

International Conference on Communication, Computing and Internet of Things (IC3IoT)

作者： Mehdi Darbandi Saeed Haghgoo Mahdi Hajiali Alireza Khabir Department of Electrical and Electronic Engineering Eastern Meditteranean University Gazimagusa Turkey Department of Computer Engineering Islamic Azad University Islamshahr Branch Iran Department of Computer Engineering and Information Technology Amirkabir University of Technology Tehran Iran Faculty of Electircal Engineering Islamic Azad university Tabriz Iran

ISBN: (纸本)9781538624609;9781538624593

关键词： Internet of Things

来源：评论

学校读者我要写书评

暂无评论

Ambient Intelligence in Automated Houses

Ambient Intelligence in Automated Houses

引用

IEEE GCC Conference & Exhibition

作者： Omar M. Abbas Amr A. Lotfy Ahmed N. Abdul-Basset Mohammed Ei-Abd Electircal and Computer Engineering Deparmtnet American University of Kuwait

ISBN: (纸本)9781538627570

Smart houses are invading nearly all modern companies under the flag of the booming field of home automation. This field is being implemented with the highest technologies to be integrated in daily life and be a non-neglect able part of everyone's life style. This field is growing at a fast pace as the field of smart devices was years ago. The main purpose of smart houses and home automation is to provide a more luxurious lifestyle to its users. However, throughout the span of the past years, there was not as much attention paid to the possibility of dangerous situations as the attention paid to providing luxurious welfare. The proposed home automation system in this work provides both luxurious and emergency features. There are six features discussed throughout this paper that shows how a house can deal with home tenant and the surrounding environment in case of emergency or dangerous situation.

关键词： Protocols Monitoring Skeleton Face recognition Home automation Security Portals

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：