检索结果-内蒙古大学图书馆

23rd International Conference on Computational Science, ICCS 2023

作者： Hasiec, Piotr Świtoński, Adam Josiński, Henryk Wojciechowski, Konrad Department of Computer Graphics Vision and Digital Systems Silesian University of Technology ul. Akademicka 16 Gliwice44-100 Poland Polish-Japanese Academy of Information Technology Aleja Legionów 2 Bytom41-902 Poland

ISBN: (纸本)9783031360206

Anomalies of gait sequences are detected on the basis of an autoencoder strategy in which input data are reconstructed from their embeddings. The denoising dense low-dimensional and sparse high-dimensional autoencoders are applied for segments of time series representing 3D rotations of the skeletal body parts. The outliers - misreconstructed time segments - are determined and classified as abnormal gait fragments. In the validation stage, motion capture data registered in the virtual reality of the Human Dynamics and Multimodal Interaction Laboratory of the Polish-Japanese Academy of Information technology equipped with Motek CAREN Extended hardware and software are used. The scenarios with audio and visual stimuli are prepared to enforce anomalies during a walk. The acquired data are labeled by a human, which results in the visible and invisible anomalies extracted. The neural network representing the autoencoder is trained using anomaly-free data and validated by the complete ones. AP (Average Precision) and ROC-AUC (Receiver Operating Characteristic – Area Under Curve) measures are calculated to assess detection performance. The influences of the number of neurons of the hidden layer, the length of the analyzed time segments and the variance of injected Gaussian noise are investigated. The obtained results, with AP = 0.46 and ROC-AUC = 0.71, are promising. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

On heterogeneity or sub-classes aspect in construction of stylometric input datasets 26

On heterogeneity or sub-classes aspect in construction of st...

引用

26th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems, KES 2022

作者： Stanczyk, Urszula Baron, Grzegorz Department of Graphics Computer Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 2A Gliwice44-100 Poland

Stylometric analysis of texts relies on learning characteristic traits of writing styles for authors. Once these patterns are discovered, they can be compared to the ones present in other text samples, to recognise their authorship. This recognition can be compromised if input datasets are prepared without taking into consideration possible stratification of the input space, leading to specific grouping of datapoints, or sub-classes within distinguished classes. The paper shows research dedicated to construction of various structures of input datasets, and combinations of such structures between train and test sets. In the research the influence of different stratification forms on the performance of selected popular classification systems was observed. To minimise the number of influencing factors, a task of authorship attribution was performed as binary classification with balanced classes. Stylometric descriptors exploited belonged to lexical and syntactic group, giving frequencies of occurrence for chosen style-markers. It resulted in real-valued attributes and these values were explored without applying discretisation, in order to avoid the possible bias of this procedure on observations. © 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://***/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the 26th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems (KES 2022)

关键词： Knowledge based systems

来源：评论

学校读者我要写书评

暂无评论

Engineering technology Capstone: Exploring the Current State

引用

Journal of Engineering technology 2024年第2期41卷 34-40页

作者： Webster, Rustin Turner, Matthew Purdue Polytechnic Institute School of Engineering Technology Purdue University mechanical engineering and computer graphics technology United States Purdue University’s School of Engineering Technology United States

The characteristics that make up the general identity of engineering technology (ET) degree programs and their graduates are well known;however, the explicit characteristics of ET capstone nationally is unknown. In other words, there are multiple gaps in the general ET capstone body of knowledge that need to be filled in order to fully understand and describe the current state. The purpose of this short paper is to provide a synopsis of 1) the current known identity of ET education, 2) existing ET capstone research, and 3) ET capstone knowledge gaps (i.e., future research areas). Internal preliminary findings comparing traditional engineering degree programs and ET capstone experiences include differences in duration, project types and sources, budgets, assessments, etc. Additionally, findings show that ET degree programs were not included in multiple large-scale national capstone-based research studies investigating current and longitudinal practices, design teaching taxonomies, and graduate career readiness. Deeper insights into ET capstone practices are needed and can help empower stakeholders (e.g., faculty, administrators, researchers, etc.) to articulate ET’s value more effectively within the broader landscape of science, technology, engineering, and mathematics (STEM), and hopefully spark interest in new research collaborations within the general capstone community in order to address identified knowledge gaps. A better understood and possibly improved ET capstone experience could help better prepare graduates to transition into the workforce by minimizing the time needed to acquire new knowledge, skills, and abilities after being hired. © 2024, American Society for Engineering Education. All rights reserved.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

Privacy-Preserving, Low-Storage, and High-Quality Image Processing in IoT and Clouds

Privacy-Preserving, Low-Storage, and High-Quality Image Proc...

引用

2023 International Conference on Intelligent Communication and computer Engineering, ICICCE 2023

作者： Zhao, Shushan Chiang, Wes Central Connecticut State University Department of Computer Electronics & Graphics Technology New BritainCT United States International Business & Strategy Brock University Department of Marketing St. CatharinesON Canada

ISBN: (纸本)9798350395617

In Internet of Things (IoT) and clouds, while many image processing tasks are outsourced to third party cloud computing platforms, image processing in encrypted domain is needed in many services for data confidentiality and privacy-preserving purposes. There are many solutions for image scaling, cropping and color correction in this context. However, most of the solutions are not compatible with each other and there are few solutions supporting all these operations. In this paper, we propose a scheme that supports image scaling, cropping, and color correction in encrypted domain with minimum storage overhead. An optional improvement solution is added for a conditional quality issue in certain circumstances. Analysis shows storage requirement is drastically reduced while performance otherwise remains unchanged. © 2023 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Automated classification of crests on pottery sherds using pattern recognition on 2D images

Automated classification of crests on pottery sherds using p...

引用

20th Eurographics Workshop on graphics and Cultural Heritage, GCH 2022

作者： Ritz, M. Santos, P. Fellner, D.W. Fraunhofer IGD Germany TU-Darmstadt Germany Graz University of Technology Institute of Computer Graphics and Knowledge Visualization Austria

ISBN: (纸本)9783038681786

Manual classification of artefacts is a labor intensive process. Based on 2D images and 3D scans of - for example - ceramic shards, we developed a pattern recognition algorithm which automatically extracts relief features for each newly recorded object and tries to automate the classification process. Based on characteristics found, previously unknown objects are automatically corelated to already classified objects of a collection exhibiting the greatest similarity. As a result, classes of artefacts form iteratively, which ultimately also corresponds to the overall goal which is the automated classification of entire collections. The greatest challenge in developing our software approach was the heterogeneity of reliefs, and in particular the fact that current machine learning approaches were out of question due to the very limited number of objects per class. This led to the implementation of an analytical approach that is capable of performing a classification based on very few artefacts. © ISVR,2023 *** rights reserved

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

M2fNet: Multi-Modal Forest Monitoring Network on Large-Scale Virtual Dataset

M2fNet: Multi-Modal Forest Monitoring Network on Large-Scale...

引用

Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW), IEEE Conference on

作者： Yawen Lu Yunhan Huang Su Sun Tansi Zhang Xuewen Zhang Songlin Fei Victor Chen Department of Computer Graphics Technology Purdue University Meta USA Department of Forestry and Natural Resources Purdue University

ISBN: (数字)9798350374490

ISBN: (纸本)9798350374506

Forest monitoring and education are key to forest protection, education and management, which is an effective way to measure the progress of a country's forest and climate commitments. Due to the lack of a large-scale wild forest monitoring benchmark, the common practice is to train the model on a common outdoor benchmark (e.g., KITTI) and evaluate it on real forest datasets (e.g., CanaTree100). However, there is a large domain gap in this setting, which makes the evaluation and deployment difficult. In this paper, we propose a new photorealistic virtual forest dataset and a multimodal transformer-based algorithm for tree detection and instance segmentation. To the best of our knowledge, it is the first time that a multimodal detection and segmentation algorithm is applied to a large-scale forest scenes. We believe that the proposed dataset and method will inspire the simulation, computer vision, education and forestry communities towards a more comprehensive multi-modal understanding.

关键词： Training computer vision Three-dimensional displays Forestry Vegetation Virtual reality Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

A SVDD Method Based on Maximum Distance between Two Centers of Spheres

引用

Chinese Journal of Electronics 2023年第1期21卷 107-111页

作者： Jinglong Fang Wanliang Wang Xingqi Wang Zhe Long Dongsheng Liang Qili Zhou School of Computer Science and Technology Zhejiang University of Technology Hangzhou China Institute of Graphics and Image Hangzhou Dianzi University Hangzhou China

On the base of the method of Support vector data description (SVDD), this paper proposes a SVDD method based on maximum distance between two centers of spheres. It applies two hyperspheres to separate two kinds of training instance and maximize distance between two centers of hyperspheres at the same time. Experimental results show that the method is effective, especially for unbalance problem, our method can get better results than all other methods.

关键词： Training Support vector machine classification Research and development

来源：评论

学校读者我要写书评

暂无评论

Real-Time 6-DoF Object Pose Estimation Network Based on 2D-3D Coordinate Correspondence 3rd

Real-Time 6-DoF Object Pose Estimation Network Based on 2D-3...

引用

3rd International Conference on Artificial Intelligence, Robotics, and Communication, ICAIRC 2023

作者： Li, Shuai Chen, Jinlong Yang, Minghao Su, Jianhua Guangxi Key Laboratory of Trusted Software Guilin University of Electronic Technology Guangxi Guilin541004 China Key Laboratory of Intelligent Processing of Computer Image and Graphics Guilin University of Electronic Technology Guangxi Guilin541004 China Beijing100190 China HuaiRou Beijing101400 China Beijing100190 China

ISBN: (纸本)9789819721993

Object 6D pose estimation is essential for a variety of applications, including autonomous driving, robotic manipulation, and automated harvesting. Due to the influence of different lighting conditions and occlusions, objects with different poses may exhibit significant variations in appearance in different views. This makes the estimation of the 6D pose of objects from a single RGB image a challenging task. This paper proposes a deep learning network capable of simultaneously detecting and estimating the 6D pose (3D rotation + 3D translation) of target objects. As establishing a large-scale dataset using real data is extremely challenging, this paper introduces an automatic collection scheme based on virtual data generated from real images. We conducted experiments on the generated astronaut dataset using our data collection scheme for 6D pose estimation. The experimental results show that the accuracy of 6D pose estimation based on the ADD metric is 96%. Therefore, this method has significant potential and can be extended to other downstream tasks. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Context-based Surface Pattern Completion of Ancient Pottery

Context-based Surface Pattern Completion of Ancient Pottery

引用

20th Eurographics Workshop on graphics and Cultural Heritage, GCH 2022

作者： Lengauer, Stefan Preiner, Reinhold Sipiran, Ivan Karl, Stephan Tinkl, Elisabeth Bustos, Benjamin Schreck, Tobias Graz University of Technology Institute of Computer Graphics and Knowledge Visualization Austria University of Chile Department of Computer Science Chile University of Graz Institute of Classics Austria

ISBN: (纸本)9783038681786

Among various ancient cultures it was common practice to adorn pottery artifacts with lavish surface decoration. While the applied painting styles, color schemes and displayed mythological content may vary greatly, the presence of simple patterns which appear in a repetitive manner can be observed across civilizations and periods. Such pattern sequences generally are arranged in a structured manner in ornament bands or columns that extend over the entire surface of the object. Due to the poor conservation state of many cultural heritage objects, parts of the surface are oftentimes badly damaged or missing altogether. Yet, if the majority of a pattern sequence is preserved, this information can be leveraged to approximate its missing parts. We present an approach that allows the fully automatic determination of the generation rule inherent to a repetitive surface pattern. Based on this generation rule and the preserved patterns from the same pattern class we propose a workflow for reconstruct missing or damaged parts of the surface painting. We evaluate our approach by applying it to a selection of pottery from ancient Peruvian and Greek cultures, showing that our automatic approach is able to handle a variety of problem cases. © ISVR,2023 *** rights reserved

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

WGAN-Based Synthetic Data Generation for Modeling Cyberattacks in Power Transmission Systems

WGAN-Based Synthetic Data Generation for Modeling Cyberattac...

引用

Innovative Approaches in Smart Technologies (ISAS), International Symposium on

作者： Spoorthy Reddy Gondhi Urmisha Reddy Janak Ashok Raja Arash Asrari Electrical and Computer Engineering Department Purdue University Northwest Hammond IN USA Computer Information Technology and Graphics Department Purdue University Northwest Hammond IN USA

ISBN: (数字)9798331540104

ISBN: (纸本)9798331540111

The integration of diverse infrastructures in modern-day power systems facilitates unauthorized access and data manipulation by adversaries, as these systems heavily rely on Information and Communication technology (ICT) for monitoring and control. A significant challenge within these power networks is the risk of operational disruptions, such as congestion and voltage instability, resulting from stealthy false data injection (FDI) cyberattacks. Different from the existing work, this paper proposes a solution by introducing a defense framework that utilizes a Wasserstein Generative Adversarial Network (WGAN) to generate synthetic data. This synthetic data closely resembles the output from actual Phasor Measurement Units (PMU) and is developed by training the WGAN with extensive real PMU datasets. Additionally, mixing synthetic and real data when sending it to the Supervisory Control and Data Acquisition (SCADA) system adds layers of complexity and obscures the data landscape for attackers, thereby hindering their ability to detect vulnerabilities and anomalies.

关键词： SCADA systems Power system stability Generative adversarial networks Phasor measurement units Data models Power system reliability Information and communication technology Reliability computer crime Synthetic data

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：