检索结果-内蒙古大学图书馆

IEEE International Conference on Intelligent computing and Information Systems (ICICIS)

作者： Omar Ahmed Sameh Fekry Abdullah Metwally Abdelrahman Abdelwahid Dina Khattab Amer Abdulelmagid Radwa Hossieny Scientific Computing Department Faculty of Computer and Information Science Ain Shams University Cairo Egypt

Due to the growing usage of DeepFake technology to produce fake photos and videos, the problem of DeepFake identification has become a serious concern in recent years. With the use of this technology, fake material that is difficult to tell apart from real content may be produced. Therefore, the development of accurate DeepFake detection algorithms is essential to identify and prevent the spread of manipulated content. However, current publicly available DeepFake detection datasets suffer from a lack of diversity, with only a few actors appearing in multiple videos. This results in an oversampled training dataset, leading to model overfitting and inadequate performance when tested on new data. The overfitting problem can cause deep neural networks to focus more on facial features than on specific traits of DeepFake content, thereby reducing the model's ability to generalize. To address this issue and broaden the training dataset's variety, we suggest applying data augmentation methods like Face Cutout and Random Erase. The Face Cutout technique randomly removes a rectangular area of an image containing the face, while the Random Erase technique randomly removes a rectangular area of the image. These techniques introduce variations in the images, making it more challenging for the model to overfit and focus on specific traits of DeepFake content. We evaluated our approach using several models and achieved significant accuracy improvements using the EfficientnetV2B0 model and Random Erase augmentation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A fully implicit method for lattice boltzmann equations

A fully implicit method for lattice boltzmann equations

引用

作者： Huang, Jizu Yang, Chao Cai, Xiao-Chuan Institute of Computational Mathematics and Scientific/Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China Institute of Software Chinese Academy of Sciences Beijing100190 China State Key Laboratory of Computer Science Beijing100190 China Department of Computer Science University of Colorado Boulder BoulderCO80309 United States

Existing approaches for solving the lattice Boltzmann equations with finite difference methods are explicit and semi-implicit;both have certain stability constraints on the time step size. In this work, a fully implicit second-order finite difference scheme is developed. We focus on a parallel, highly scalable, Newton-Krylov-RAS algorithm for the solution of a large sparse nonlinear system of equations arising at each time step. Here, RAS is a restricted additive Schwarz preconditioner based on a first-order spatial discretization. We show numerically that by using the fully implicit method the time step size is no longer constrained by the CFL condition, and the Newton-Krylov- RAS algorithm is scalable on a supercomputer with more than ten thousand processors. Moreover, to calculate the steady state solution we investigate an adaptive time stepping strategy. The total compute time required by the implicit method with adaptive time stepping is much smaller than that of an explicit method for several test cases. © 2015 Society for Industrial and Applied Mathematics.

关键词： Finite difference method

来源：评论

学校读者我要写书评

暂无评论

The generic graph component library 99

The generic graph component library

引用

Proceedings of the 14th ACM SIGPLAN conference on Object-oriented programming, systems, languages, and applications

作者： Lie-Quan Lee Jeremy G. Siek Andrew Lumsdaine Laboratory for Scientific Computing Department of Computer Science and Engineering University of Notre Dame Notre Dame IN

ISBN: (纸本)9781581132380

In this paper we present the Generic Graph Component Library (GGCL), a generic programming framework for graph data structures and graph algorithms. Following the theme of the Standard Template Library (STL), the graph algorithms in GGCL do not depend on the particular data structures upon which they operate, meaning a single algorithm can operate on arbitrary concrete representations of graphs. To attain this type of flexibility for graph data structures, which are more complicated than the containers in STL, we introduce several concepts to form the generic interface between the algorithms and the data structures, namely, Vertex, Edge, Visitor, and Decorator. We describe the principal abstractions comprising the GGCL, the algorithms and data structures that it provides, and provide examples that demonstrate the use of GGCL to implement some common graph algorithms. Performance results are presented which demonstrate that the use of novel lightweight implementation techniques and static polymorphism in GGCL results in code which is significantly more efficient than similar libraries written using the object-oriented paradigm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multiscale Adaptive Correlation Method for Ultrasound Speckle Image Velocimetry

Multiscale Adaptive Correlation Method for Ultrasound Speckl...

引用

IEEE International Ultrasonics Symposium

作者： Ecaterina Bodnariuc Christoph Schnorr Interdisciplinary Center for Scientific Computing (IWR) Department of Mathematics and Computer Science Heidelberg University Germany

In this paper, we introduce and evaluate a multiscale correlation-based image velocimetry approach to the estimation of in-plane displacements of ultrasound speckles. The method replaces the conventional rectangular interrogation window by a smooth Gaussian weight function and locally adapts its shape and orientation to estimated velocities by evaluating a corresponding criterion. Furthermore, displacements are continuously refined which improves the sub-pixel accuracy of the computed velocities. Experimental results are reported that we obtained by applying the method to the SA-VFI challenge datasets.

关键词： Microsoft Windows Ultrasonic imaging Speckle Correlation Shape Media Image color analysis

来源：评论

学校读者我要写书评

暂无评论

A filter representation for reduced order models

A filter representation for reduced order models

引用

IEEE Workshop on Signal Propagation on Interconnects (SPI)

作者： P.J. Heres W.H.A. Schilders Department of Mathematics and Computer Science Scientific Computing Group Eindhovan University of Technology Eindhoven Netherlands

来源：评论

学校读者我要写书评

暂无评论

Using graphs for fast error term approximation of time-varying datasets

Using graphs for fast error term approximation of time-varyi...

引用

5th Joint Eurographics - IEEE TCVG Symposium on Visualization, VisSym 2003

作者： Nuber, C. LaMar, E.C. Pascucci, V. Hamann, B. Joy, K.I. Center for Image Processing and Integrated Computing Department of Computer Science University of California Davis DavisCA95616-8562 United States Center for Applied Scientific Computing Lawrence Livermore National Laboratory LivermoreCA94550-9234 United States

ISBN: (纸本)3905673010

We present a method for the efficient computation and storage of approximations of error tables used for error estimation of a region between different time steps in time-varying datasets. The error between two time steps is defined as the distance between the data of these time steps. Error tables are used to look up the error between different time steps of a time-varying dataset, especially when run time error computation is expensive. However, even the generation of error tables itself can be expensive. For n time steps, the exact error look-up table (which stores the error values for all pairs of time steps in a matrix) has a memory complexity and pre-processing time complexity of O(n2), and O(1) for error retrieval. Our approximate error look-up table approach uses trees, where the leaf nodes represent original time steps, and interior nodes contain an average (or best-representative) of the children nodes. The error computed on an edge of a tree describes the distance between the two nodes on that edge. Evaluating the error between two different time steps requires traversing a path between the two leaf nodes, and accumulating the errors on the traversed edges. For n time steps, this scheme has a memory complexity and pre-processing time complexity of O(nlog(n)), a significant improvement over the exact scheme;the error retrieval complexity is O(log(n)). As we do not need to calculate all possible n2 error terms, our approach is a fast way to generate the approximation. © 2003 Eurographics Association. All rights reserved.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

Mobile Money Phishing Cybercrimes: Vulnerabilities, Taxonomies, Characterization from an Investigation in Cameroon 14th

Mobile Money Phishing Cybercrimes: Vulnerabilities, Taxonomi...

引用

14th EAI International Conference on Towards new e-Infrastructure and e-Services for Developing Countries, AFRICOMM 2022

作者： Nzeket Njoya, Alima Tchakounté, Franklin Atemkeng, Marcellin Udagepola, Kalum Priyanath Bassolé, Didier Department of Mathematics and Computer Science Faculty of Science University of Ngaoundere Ngaoundere Cameroon Ngaoundere Cameroon Department of Mathematics Rhodes University Grahamstown6140 South Africa Department of Information and Computing Sciences Scientific Research Development Institute of Technology Loganlea Australia Laboratory of Mathematics and Computer Science University of Joseph Ki-Zerbo Ouagadougou Burkina Faso

ISBN: (纸本)9783031348952

Mobile Money (MM) technologies are popular in developing countries where people are unbanked, and they are exploited as means of financial transactions in the economy. Sophisticated cyber-phishing techniques successfully target MM accounts. Related countermeasures are rare and the existing ones are so technical that people without minimal knowledge cannot be helped. Making the knowledge around cybercrime facts is therefore relevant since it provides a good basis for further technical research. In this vein, this paper dissects phishing cybercrime strategies observed within the Cameroonian cyberspace. We provide identified vulnerabilities, process design of generic MM attack, taxonomies of attacks, and classification of the latter based on criteria. Findings about commonalities and dissimilarities reveal two aspects to really consider when designing solutions: emotion and interactions. © 2023, ICST Institute for computer sciences, Social Informatics and Telecommunications Engineering.

关键词： Taxonomies

来源：评论

学校读者我要写书评

暂无评论

Handwritten chemical formulas classification model using deep transfer convolutional neural networks 2

Handwritten chemical formulas classification model using dee...

引用

2nd IEEE International Conference on Electronic Engineering, ICEEM 2021

作者： Hagag, Ahmed Omara, Ibrahim Alfarra, Ahmed N. K. Mekawy, Fahd Benha University Faculty of Computers and Artificial Intelligence Department of Scientific Computing Benha13518 Egypt University Menoufia Faculty of Science Menoufia Department of Mathematics and Computer Science Menoufia32511 Egypt Harbin Institute of Technology School of Management Harbin150001 China

ISBN: (纸本)9781665418423

With the spread of the COVID19 pandemic, blended learning has become one of the most used methods in educational organizations such as universities, community colleges, and schools. In blended learning, the students' practical activities are done in more than one way, including simulation software and the place of study. For chemical experiment programs, the classification of handwritten chemical formulas plays an important role in determining the simulation software's efficiency. Accordingly, in this study, we propose a model for handwritten chemical formula classification. First, this paper describes a handwritten chemical formulas dataset that contains eight classes (HCFD8). Second, convolutional neural networks (CNNs) with pre-trained weights are used as a deep feature extractor to extract features from the images. Third, due to limited training images per class, the proposed model uses data augmentation techniques to expand the training images. Then, an enhanced multilayer perceptron (EMLP) strategy is used to classify the image. Finally, we provide a performance analysis of typical deep learning approaches on HCFD8, which shows that the proposed model performs good accuracy results. © 2021 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

One variant of a （2 ＋ 1）-dimensional Volterra system and its （1 ＋ 1）-dimensional reduction

引用

Frontiers of Mathematics in China 2013年第5期8卷 1085-1097页

作者： Yingnan ZHANG Yi HE Hon-Wah TAM LSEC Institute of Computational Mathematics and Scientific Engineering ComputingAcademy of Mathematics and System Sciences Chinese Academy of SciencesBeijing 100190 China University of Chinese Academy of Sciences Beijing 100049 China Wuhan Institute of Physics and Mathematics Chinese Academy of Sciences Wuhan 430071 China Department of Computer Science Hong Kong Baptist University Hong Kong China

A new system is generated from a multi-linear form of a （2＋1）- dimensional Volterra system. Though the system is only partially integrable and needs additional conditions to possess two-soliton solutions, its （1＋1）- dimensional reduction gives an integrable equation which has been studied via reduction skills. Here, we give this （1＋1）-dimensional reduction a simple bilinear form, from which a Backlund transformation is derived and the corresponding nonlinear superposition formula is built.

关键词： Integrability soliton solution Bgcklund transformation （BT）,nonlinear superposition formula

来源：评论

学校读者我要写书评

暂无评论

Parallel multilevel block ILU preconditioning techniques for large sparse linear systems

Parallel multilevel block ILU preconditioning techniques for...

引用

International Symposium on Parallel and Distributed Processing (IPDPS)

作者： Chi Shen Jun Zhang Kai Wang Laboratory for High Performance Scientific Computing and Computer Simulation Department of Computer Science University of Kentucky Lexington KY USA

ISBN: (纸本)0769519261

We present a class of parallel preconditioning strategies built on a multilevel block incomplete LU (ILU)factorization technique to solve large sparse linear systems on distributed memory parallel computers. The preconditioners are constructed by using the concept of block independent sets. Two algorithms for constructing block independent sets of a distributed sparse matrix are proposed. We compare a few implementations of the parallel multilevel ILU preconditioners with different block independent set algorithms and different coarse level solution strategies. We also use some diagonal thresholding and perturbation strategies to enhance factorization stability. Numerical experiments indicate that our parallel multilevel block ILU preconditioners are robust and efficient.

关键词： Linear systems Sparse matrices Concurrent computing Distributed computing Robustness High performance computing Testing Robust stability Laboratories scientific computing

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：