检索结果-内蒙古大学图书馆

34th International Conference on software engineering and Knowledge engineering, SEKE 2022

作者： Wang, Xi Miao, Weikai School of Computer Engineering and Science Shanghai University China Shanghai Key Laboratory of Computer Software Testing and Evaluating Shanghai201114 China Software Engineering Institute East China Normal University China

ISBN: (纸本)1891706543

The rapid development of machine learning (ML) systems has raised many concerns over their quality. Due to the inherent complexity and uncertainty, most of the traditional quality assurance techniques have been challenged, including requirements specification. Current strategies mainly focus on model extraction from existing neural networks to improve interpretability and facilitate system analysis, but failing to include user expectations on the system. To handle the problem, this paper proposes a specification framework for ML requirements where each ML system is regarded as a set of snapshot systems along the evolvement process. There are 3 layers in the framework and the hierarchy indicates that higher-level models need to be built based on lower-level ones. The bottom layer consists of meta snapshot model and meta data model serving as the meta models for snapshot systems and data requirements respectively. The middle layer is for snapshot models each describing a snapshot system through relations between its outputs produced with different inputs. The top layer is a learning model capturing the evolvement process by transitions among snapshot models. These transitions are activated by data models instantiated from meta data model. We adopt the specification of a self-driving system to illustrate the framework. © 2022 Knowledge Systems Institute Graduate school. All rights reserved.

关键词： Metadata

来源：评论

学校读者我要写书评

暂无评论

ATMLP: Attention and Time Series MLP for Fall Detection

引用

IAENG International Journal of Applied Mathematics 2023年第1期53卷 1页

作者： Zhang, Huan Cui, Wenhua Shi, Tianwei Tao, Ye Zhang, Jianfeng School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

Addressing the problem that the human body is difficult to be correctly identified in the process of falling, this study proposes an Attention Mechanism and Time Series MLP (AT-MLP) model. To realize the capture of body features during the state and process of the fall of the human body, the time series is combined in the submodule. Thus, the subspace Angle Time Series algorithm (ATS) and the subspace Angle Difference-value Time Series algorithm (ADTS) are designed. The AT-MLP stage composed of Convolutional Block Attention Module (CBAM) and MLP blocks is constructed in the model to strengthen the relation between the time and space dimensions. Comparing with other methods on the Le2i Fall detection Dataset (LFDD) to verify the effectiveness of the AT-MLP model, the experimental results show that the AT-MLP model effectively improves the detection efficiency of human falls and has strong generalization ability. © 2023,IAENG International Journal of Applied Mathematics. All Rights Reserved.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

High-Performance GPU based Planetary-Scale Terrain Visualization 2022

High-Performance GPU based Planetary-Scale Terrain Visualiza...

引用

5th International Conference on Image and Graphics Processing, ICIGP 2022

作者： Masood, Zafar Zheng, Jiangbin Ahmad, Idrees Irfan, Muhammad Saeed, Amir School of Software Northwestern Polytechnical University Xi’an China Department of Computer Science and Software Engineering International Islamic University Islamabad Pakistan School of Automation and Electrical Engineering Beihang University China

ISBN: (纸本)9781450395465

High-performance visualization of planetary-scale multi-resolution terrain for high-fidelity display systems is a challenging task. Real-world terrain visualization requires real-time modeling and rendering of the large-scale multi-resolution model. This work proposes a novel method for high-performance multi-resolution terrain modeling and rendering using hardware tessellation. The proposed method submits patch primitives using geometry instancing for high-performance. Patch Level-of-detail (LOD) is simplified using a view-dependent simplification algorithm with error control. The proposed simplification algorithm simplifies the model using viewer information and local height variation for optimum triangulation. Patch simplified LOD is tessellated utilizing a fixed-function tessellator. Height grid-aligned post-tessellated vertices are displaced in 3D to construct accurate high-frequency surface details. Finally, multi-resolution satellite imagery and normal maps are utilized for real-world features display. We evaluated the proposed method for high display resolutions and achieved high rendering rate. The proposed method results are compared with clipmap-based and hardware tessellation based methods. © 2022 ACM.

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

Vehicle And Pedestrian Detection Algorithm Based on Improved YOLOv5

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2023年第4期50卷 1401页

作者： Sun, Jiuhan Wang, Zhifeng School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

As urbanization progresses, urban road congestion has intensified, highlighting the need for effective vehicle and pedestrian detection as a cornerstone of public safety transportation. This area holds significant relevance in video surveillance and public safety domains. Despite its importance, achieving precise vehicle and pedestrian detection in complex road environments remains challenging. This paper presents a vehicle-pedestrian detection algorithm based on the improved YOLOv5. Key modifications include the integration of a small target detection layer and alterations to the feature pyramid using the feature fusion technique inherent to the weighted Bidirectional Feature Pyramid Network (BIFPN). This ensures efficient multi-scale feature fusion. A coordinated attention mechanism is introduced to preserve accurate target location data. Furthermore, the paper incorporates the SIOU metric to refine the localization loss function, bolstering both speed and edge regression accuracy. Experimental outcomes indicate that our improved YOLOv5 algorithm augments detection accuracy by 1.9% and achieves a detection speed of 67 FPS, which surpasses many competing target detection algorithms. © 2023, International Association of Engineers. All rights reserved.

关键词： Vehicles

来源：评论

学校读者我要写书评

暂无评论

Building the metaverse using digital twins at all scales,states,and relations

引用

Virtual Reality & Intelligent Hardware 2022年第6期4卷 459-470页

作者： Zhihan LV Shuxuan XIE Yuxi LI M.Shamim HOSSAIN Abdulmotaleb El SADDIK Department of Game Design Faculty of ArtsUppsala UniversitySweden College of Computer Science and Technology Qingdao UniversityQingdaoChina Department of Software Engineering College of Computer and Information SciencesKing Saud UniversityRiyadhSaudi Arabia School of Electrical Engineering and Computer Science(EECS) University of OttawaOttawaOntarioCanada

Developments in new-generation information technology have enabled Digital Twins to reshape the physical world into a virtual digital space and provide technical support for constructing the *** objects can be at the micro-,meso-,or *** Metaverse is a complex collection of solid,liquid,gaseous,plasma,and other uncertain ***,the Metaverse integrates tangibles with social relations,such as interpersonal(friends,partners,and family)and social relations(ethics,morality,and law).This review introduces some principles and laws,such as broken windows theory,small-world phenomenon,survivor bias,and herd behavior,for constructing a Digital Twins model for social ***,from multiple perspectives,this article reviews mappings of tangible and intangible real-world objects to the Metaverse using the Digital Twins model.

关键词： Digital twins Metaverse Sensor

来源：评论

学校读者我要写书评

暂无评论

Lightweight Vehicle Detection Based on Improved Yolox-nano

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2023年第1期50卷 152页

作者： Jing, XiuPing Tian, Ying School of Computer Science and Software Engineering University of Science and Technology Liaoning Liaoning 114051 China

Vehicle detection based on deep learning plays a vital role in various fields, such as autopilot and intelligent transportation. Moreover, it presents a major development direction to computer vision in recent years. Lightweight vehicle detection aims to reduce the burden of computer storage and computing, including exploration of network structure and computing efficiency. But it is difficult to cope with complex and changeable traffic scenes. The urgent problem is to enhance lightweight network performance while maintaining inference speed. In this context, this study proposes an improved Yolox-nano to alleviate the above-mentioned problems by splitting the vanilla convolution unit into two parts and utilizing fewer convolution kernels to generate several feature maps. Subsequently, several linear transformations are further applied to generate cheap feature maps efficiently. The cheap convolutional unit Ghost Module can control feature map ratio of vanilla convolution and linear transformation. The activation function in Ghost Module is improved by SiLU, which has a more powerful nonlinear ability. In addition, the convolution and the batch normalization in Ghost Module can be fused in the pre-processing stage to speed up inference time. Experimental results show that the improved network provides 4.9% and 3.1% mAP on Pascal VOC and MS COCO vehicle datasets. The improved network enhances the performance and maintains an efficient inference speed, and detection results become further competitive using Yolox-nano © 2023, IAENG International Journal of computer *** Rights Reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Cyber Security Threats on Multifunctional Devices and Mitigation Techniques 2

Cyber Security Threats on Multifunctional Devices and Mitiga...

引用

2nd International Conference on Business Analytics for Technology and Security, ICBATS 2023

作者： Al-Dmour, Nidal A. Farooq, Umer Sarwar, Irfan Iqbal, Muhammad Waseem Aqeel, Muhammad Khan, Wasim Ahmad Al Hamadi, Hussam Mutah University College of Engineering Department of Computer Engineering Jordan LGU Department of Computer Science Lahore Pakistan Superior University Department of Computer Science Lahore54000 Pakistan Superior University Department of Software Engineering Lahore54000 Pakistan NCBA&E Lahore School of Computer Science Pakistan Applied Science Research Center Applied Science Private University Amman11937 Jordan University of Dubai College of Engineering and IT PO Box 14143 United Arab Emirates

ISBN: (纸本)9798350335644

The risks discussed in this document are related to smart printers/ fax/ photocopiers/ or e-mail-enabled smart devices, also known as "document centers,". These devices can be purchased outright or rented every month. When an item reaches its end of life, it is either donated to a non-profit organization or returned to the manufacturer (OEM) or supplier at the end of a lease arrangement. These network-interconnected devices have an inherent susceptibility that most IT operations teams or information security professionals are unaware of. This effort will raise awareness of the dangers of losing a company's data. This discussion will focus on reported vulnerabilities, previous discussions, and given the author's view on the potential danger. Create instructions for potential users on how to use and disassemble these devices safely to prevent losing sensitive data. Lastly, the technological structure of two potential attack vectors against a document center of an organization will be demonstrated in this study. The study was conducted in government as well as the private sector, medium-sized enterprises (300 employees of each entity) settings, outside of normal business hours, with special authorizations and engagement, limiting the scope to a minimal set of devices. © 2023 IEEE.

关键词： Sensitive data

来源：评论

学校读者我要写书评

暂无评论

Multiplex Networks and Pan-Cancer Multiomics-Based Driver Gene Identification Using Graph Neural Networks

引用

Big Data Mining and Analytics 2024年第4期7卷 1262-1272页

作者： Xingyi Li Junming Li Jun Hao Xingyu Liao Min Li Xuequn Shang School of Computer Science Northwestern Polytechnical UniversityXi’an 710072Chinaand with Research&Development Institute of Northwestern Polytechnical University in ShenzhenShenzhen 518063China School of Software Northwestern Polytechnical UniversityXi’an 710072Chinaand with Research&Development Institute of Northwestern Polytechnical University in ShenzhenShenzhen 518063China School of Computer Science Northwestern Polytechnical UniversityXi’an 710072China School of Computer Science and Engineering Central South UniversityChangsha 410083China

Identifying cancer driver genes has paramount significance in elucidating the intricate mechanisms underlying cancer development,progression,and therapeutic *** omics data and interactome networks provided by numerous extensive databases enable the application of graph deep learning techniques that incorporate network structures into the deep learning ***,most existing models primarily focus on individual network,inevitably neglecting the incompleteness and noise of ***,samples with imbalanced classes in driver gene identification hamper the performance of *** address this,we propose a novel deep learning framework MMGN,which integrates multiplex networks and pan-cancer multiomics data using graph neural networks combined with negative sample inference to discover cancer driver genes,which not only enhances gene feature learning based on the mutual information and the consensus regularizer,but also achieves balanced class of positive and negative samples for model *** reliability of MMGN has been verified by the Area Under the Receiver Operating Characteristic curves(AUROC)and the Area Under the Precision-Recall Curves(AUPRC).We believe MMGN has the potential to provide new prospects in precision oncology and may find broader applications in predicting biomarkers for other intricate diseases.

关键词： cancer driver gene multiplex networks pan-cancer multiomics data graph neural networks negative sample inference

来源：评论

学校读者我要写书评

暂无评论

Framework for improving software Requirements via Automated Behavior-Driven Development: A Priority-Based Approach*

Framework for improving Software Requirements via Automated ...

引用

International Conference on Communication Technologies (ComTech)

作者： Zainab Yousaf Muhammad Sohail Javed Iqbal Asad Ullah Sumet Mehta Department of Computer Science Software Engineering National University of Science and Technology Islamabad Pakistan School of Computer Science and Communication Engineering Jiangsu University Zhenjiang Jiangsu China

ISBN: (数字)9798331515331

ISBN: (纸本)9798331515348

A novel framework for automating behavior-driven development (BDD) and improving software requirements is presented in this paper using a priority-based approach. Through the integration of BDD concepts with automated prioritization techniques, the framework seeks to address the difficulties associated with software demand prioritization. The main goal is to make sure that important features are addressed first by coordinating software development efforts with business priorities. The Priority Predictor, a crucial element that uses BDD scenarios to rank software needs, was designed and implemented as part of the framework’s development. The framework is simple to use in development processes since it interfaces smoothly with popular BDD tools like Cucumber.

关键词： software Communications technology software development management Business

来源：评论

学校读者我要写书评

暂无评论

CRAYM: neural field optimization via camera RAY matching 24

CRAYM: neural field optimization via camera RAY matching

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Liqiang Lin Wenpeng Wu Chi-Wing Fu Hao Zhang Hui Huang College of Computer Science and Software Engineering Shenzhen University Department of Computer Science and Engineering The Chinese University of Hong Kong School of Computing Science Simon Fraser University

ISBN: (纸本)9798331314385

We introduce camera ray matching (CRAYM) into the joint optimization of camera poses and neural fields from multi-view images. The optimized field, referred to as a feature volume, can be "probed" by the camera rays for novel view synthesis (NVS) and 3D geometry reconstruction. One key reason for matching camera rays, instead of pixels as in prior works, is that the camera rays can be parameterized by the feature volume to carry both geometric and photometric information. Multi-view consistencies involving the camera rays and scene rendering can be naturally integrated into the joint optimization and network training, to impose physically meaningful constraints to improve the final quality of both the geometric reconstruction and photorealistic rendering. We formulate our per-ray optimization and matched ray coherence by focusing on camera rays passing through keypoints in the input images to elevate both the efficiency and accuracy of scene correspondences. Accumulated ray features along the feature volume provide a means to discount the coherence constraint amid erroneous ray matching. We demonstrate the effectiveness of CRAYM for both NVS and geometry reconstruction, over dense- or sparse-view settings, with qualitative and quantitative comparisons to state-of-the-art alternatives.

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：