检索结果-内蒙古大学图书馆

International Conference on Signal Processing and Information Security (ICSPIS)

作者： Zhanlin Ji Shengnan Hao Jianhua Pang Ivan Ganchev College of Mathematics and Computer Science Zhejiang A&F University Hangzhou China Telecommunications Research Centre University of Limerick Limerick Ireland Department of Artificial Intelligence North China University of Science and Technology Tangshan China Department of Computer Systems University of Plovdiv “Paisii Hilendarski” Plovdiv Bulgaria Institute of Mathematics and Informatics—Bulgarian Academy of Sciences Sofia Bulgaria

ISBN: (数字)9798350368673

ISBN: (纸本)9798350368680

Lung cancer, as one of the diseases with the highest incidence and mortality rates in the world, has always been a big challenge that humans are constantly striving to address. Pulmonary nodules, as one of the early manifestations of lung cancer, constitute a breakthrough in early detection of this disease and its prompt treatment. Currently, significant progress has been made in object detection models based on deep learning in the context of increasingly scarce medical resources. To make pulmonary nodule detection more efficient and accurate, this paper proposes two novel models, based on the Single Shot MultiBox Detector (SSD) model. The proposed models adapt to the detection of small pulmonary nodules by changing the size of the feature map of the input prediction module. In addition, the prediction frame is modified to make greater use of the shallow feature layer. A group of up-sampling modules is added by increasing asymmetric convolution and a feature pyramid network (FPN) mechanism is used for improving the detection performance. The experimental results, obtained on the LUNA16 dataset, demonstrate that the proposed models achieve great performance improvement compared to the baseline (SSD).

关键词： Adaptation models Convolution Lungs Computed tomography Computational modeling Biological system modeling Lung cancer Feature extraction Tumors Diseases

来源：评论

学校读者我要写书评

暂无评论

Impact of Channel Numbers and Training Epochs on U-Net’s Polyp Segmentation Performance

Impact of Channel Numbers and Training Epochs on U-Net’s Po...

引用

International Conference on Signal Processing and Information Security (ICSPIS)

作者： Zhanlin Ji Ivan Ganchev College of Mathematics and Computer Science Zhejiang A&F University Hangzhou China Telecommunications Research Centre University of Limerick Limerick Ireland University of Plovdiv “Paisii Hilendarski” Plovdiv Bulgaria Institute of Mathematics and Informatics- Bulgarian Academy of Sciences Sofia Bulgaria

ISBN: (数字)9798350368673

ISBN: (纸本)9798350368680

Medical image segmentation plays an extremely crucial role in the modern field of medicine. It is a process that involves the precise delineation and labeling of medical images through computer technology. However, during the training of segmentation models, insufficient computational power is often a challenge. Despite having superior models, the lack of hardware resources may hinder the widespread implementation of these models. Therefore, in this paper, the U-Net model is used as a case study to investigate the impact of the number of channels and training epochs on the segmentation performance. The experimental results show that by reducing the number of channels in the U-Net model and increasing the number of training epochs, the model’s parameter counts can be successfully decreased, while also achieving satisfactory segmentation results.

关键词： Training Image segmentation Computational modeling Information security Signal processing Hardware Labeling Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Blockchain-Based Decentralized Authentication Model for IoT-Based E-Learning and Educational Environments

引用

computers, Materials & Continua 2023年第5期75卷 3133-3158页

作者： Osama A.Khashan Sultan Alamri Waleed Alomoush Mutasem K.Alsmadi Samer Atawneh Usama Mir Research and Innovation Centers Rabdan AcademyP.O.Box 114646Abu DhabiUnited Arab Emirates College of Computing and Informatics Saudi Electronic UniversityRiyadh13316Saudi Arabia School of Information Technology Skyline University CollegeP.O.Box 1797SharjahUnited Arab Emirates Department of MIS College of Applied Studies and Community ServicesImam Abdulrahman Bin Faisal UniversityP.O.Box 1982DammamSaudi Arabia Department of Computer Science University of WindsorWindsorN9J3Y1Canada

In recent times,technology has advanced significantly and is currently being integrated into educational environments to facilitate distance learning and interaction between *** the Internet of Things(IoT)into education can facilitate the teaching and learning process and expand the context in which students ***,learning data is very sensitive and must be protected when transmitted over the network or stored in data ***,the identity and the authenticity of interacting students,instructors,and staff need to be verified to mitigate the impact of ***,most of the current security and authentication schemes are centralized,relying on trusted third-party cloud servers,to facilitate continuous secure *** addition,most of these schemes are resourceintensive;thus,security and efficiency issues arise when heterogeneous and resource-limited IoT devices are being *** this paper,we propose a blockchain-based architecture that accurately identifies and authenticates learners and their IoT devices in a decentralized manner and prevents the unauthorized modification of stored learning records in a distributed university *** allows students and instructors to easily migrate to and join multiple universities within the network using their identity without the need for user *** proposed architecture was tested using a simulation tool,and measured to evaluate its *** simulation results demonstrate the ability of the proposed architecture to significantly increase the throughput of learning transactions(40%),reduce the communication overhead and response time(26%),improve authentication efficiency(27%),and reduce the IoT power consumption(35%)compared to the centralized authentication *** addition,the security analysis proves the effectiveness of the proposed architecture in resisting various attacks and ensuring the security requirements of learning data in the unive

关键词： Blockchain decentralized authentication Internet of Things(IoT) E-learning IoT security

来源：评论

学校读者我要写书评

暂无评论

A Generalized Hybrid Hoare Logic

arXiv

引用

arXiv 2023年

作者： Zhan, Naijun Jin, Xiangyu Zhan, Bohua Wang, Shuling Guelev, Dimitar School of Computer Science Peking University Institute of Software Chinese Academy of Sciences Beijing China State Key Lab. of Computer Science Institute of Software Chinese Academy of Sciences University of CAS Beijing China Huawei Technologies Co. Ltd. Beijing China National Key Laboratory of Space Integrated Information System Institute of Software Chinese Academy of Sciences University of CAS Beijing China Institute of Mathematics and Informatics Bulgarian Academy of Sciences Sofia Bulgaria

Deductive verification of hybrid systems (HSs) increasingly attracts more attention in recent years because of its power and scalability, where a powerful specification logic for HSs is the cornerstone. Often, HSs are naturally modelled by concurrent processes that communicate with each other. However, existing specification logics cannot easily handle such models. In this paper, we present a specification logic and proof system for Hybrid Communicating Sequential Processes (HCSP), that extends CSP with ordinary differential equations (ODE) and interrupts to model interactions between continuous and discrete evolution. Because it includes a rich set of algebraic operators, complicated hybrid systems can be easily modelled in an algebra-like compositional way in HCSP. Our logic can be seen as a generalization and simplification of existing hybrid Hoare logics (HHL) based on duration calculus (DC), as well as a conservative extension of existing Hoare logics for concurrent programs. Its assertion logic is the first-order theory of differential equations (FOD), together with assertions about traces recording communications, readiness, and continuous *** prove continuous relative completeness of the logic w.r.t. FOD, as well as discrete relative completeness in the sense that continuous behaviour can be arbitrarily approximated by discretization. Finally, we implement the above logic in Isabelle/HOL, and apply it to verify two case studies to illustrate the power and scalability of our logic. Copyright © 2023, The Authors. All rights reserved.

关键词： Scalability

来源：评论

学校读者我要写书评

暂无评论

STEM Educational Kit for Assistance of Individuals with Special Needs

STEM Educational Kit for Assistance of Individuals with Spec...

引用

Automatics and informatics (ICAI), International Conference

作者： Todor Todorov Pajtim Vela St. Cyril and St. Methodius University of Veliko Tamovo Bulgaria Institute of Mathematics and Informatics Bulgarian Academy of Sciences Sofia Bulgaria Faculty of Computer Science University of Kadri Zeka Gjilan Kosovo

The paper presents a practical application of microcontrollers and sensors to enhance learning system and create equipment that can help individuals with special needs. Throughout this study, 26 high school students volunteered as research participants. To achieve satisfactory results, several interviews were conducted to identify some of the service needs for people with special needs. As a result, a prototype was created, in which several components were integrated to finish the equipment. Finally, is made an analysis of test results about the device performance and satisfaction of the students and teachers involved in the process.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Lookahead Exploration with Neural Radiance Representation for Continuous Vision-Language Navigation

Lookahead Exploration with Neural Radiance Representation fo...

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Zihan Wang L. Xiangyang Jiahao Yang Yeqi Liu H. Junjie Ming Jiang Shuqiang Jiang Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Department of Computer Science University of Wisconsin Madison WI USA Department of Biostatistics and Medical Informatics University of Wisconsin Madison WI USA Department of Human-centered Computing Indiana University Indianapolis IN USA

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Vision-and-language navigation (VLN) enables the agent to navigate to a remote location following the natural language instruction in 3D environments. At each navigation step, the agent selects from possible candidate locations and then makes the move. For better navigation planning, the lookahead exploration strategy aims to effectively evaluate the agent's next action by accurately anticipating the future environment of candidate locations. To this end, some existing works predict RGB images for future environments, while this strategy suffers from image distortion and high computational cost. To address these issues, we propose the pre-trained hierarchical neural radiance representation model (HNR) to produce multi-level semantic features for future environments, which are more robust and efficient than pixel-wise RGB reconstruction. Furthermore, with the predicted future environmental representations, our lookahead VLN model is able to construct the navigable future path tree and select the optimal path via efficient parallel evaluation. Extensive experiments on the VLN-CE datasets confirm the effectiveness of our method. The code is available at https://***/MrZihan/HNR-VLN

关键词： Solid modeling Three-dimensional displays Navigation Computational modeling Semantics Predictive models Rendering (computer graphics)

来源：评论

学校读者我要写书评

暂无评论

Human-imperceptible, Machine-recognizable Images

arXiv

引用

arXiv 2023年

作者： Hao, Fusheng He, Fengxiang Wang, Yikai Wu, Fuxiang Zhang, Jing Cheng, Jun Tao, Dacheng Shenzhen Institute of Advanced Technology CAS China The Chinese University of Hong Kong Hong Kong AIAI School of Informatics University of Edinburgh United Kingdom JD Explore Academy *** Inc China Deptment of Computer Science and Technology Tsinghua University China School of Computer Science The University of Sydney Australia

Massive human-related data is collected to train neural networks for computer vision tasks. A major conflict is exposed relating to software engineers between better developing AI systems and distancing from the sensitive training data. To reconcile this conflict, this paper proposes an efficient privacy-preserving learning paradigm, where images are first encrypted to become "human-imperceptible, machine-recognizable" via one of the two encryption strategies: (1) random shuffling to a set of equally-sized patches and (2) mixing-up sub-patches of the images. Then, minimal adaptations are made to vision transformer to enable it to learn on the encrypted images for vision tasks, including image classification and object detection. Extensive experiments on ImageNet and COCO show that the proposed paradigm achieves comparable accuracy with the competitive methods. Decrypting the encrypted images requires solving an NP-hard jigsaw puzzle or an ill-posed inverse problem, which is empirically shown intractable to be recovered by various attackers, including the powerful vision transformer-based attacker. We thus show that the proposed paradigm can ensure the encrypted images have become human-imperceptible while preserving machine-recognizable information. The code is available at https://***/FushengHao/PrivacyPreservingML. © 2023, CC BY-NC-ND.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Graph Burning: Bounds and Hardness

arXiv

引用

arXiv 2024年

作者： Antony, Dhanyamol Das, Anita Gosavi, Shirish Jacob, Dalu Kulamarva, Shashanka Department of Computer Science and Automation Indian Institute of Science Bengaluru India School of Data Science Indian Institute of Science Education and Research Thiruvananthapuram India Department of Mathematics Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India Department of Mathematics Indian Institute of Technology Delhi India School of Informatics Kyoto University Kyoto Japan

For an undirected graph G, graph burning is defined as follows: at step t = 0 all vertices in G are unburned. At each step t ≥ 1, one new unburned vertex is selected to burn until we exhaust all the vertices. If a vertex is burned at step t, then all its unburned neighbors are burned in step t + 1, and the process continues until there are no unburned vertices in G. The burning number of a graph G, denoted by b(G), is the minimum number of steps required to burn all the vertices of G. The Burning Number problem asks whether the burning number of an input graph G is at most k or not. In this paper, we study the Burning Number problem both from an algorithmic and a structural point of view. The Burning Number problem is known to be NP-complete for trees with maximum degree at most three and interval graphs. Here, we prove that this problem is NP-complete even when restricted to connected cubic graphs and connected proper interval graphs. The well-known burning number conjecture asserts that all the vertices of a graph of order n can be burned in ⌈√n ⌉ steps. In line with this conjecture, upper and lower bounds of b(G) are well-studied for various graph classes. Here, we provide an improved upper bound for the burning number of connected Pk-free graphs and show that the bound is tight up to an additive constant 1. Finally, we study two variants of the problem, namely edge burning (only edges are burned) and total burning (both vertices and edges are burned). In particular, we establish their relationship with the burning number problem and evaluate the algorithmic complexity of these *** Codes 05C38, 05C05, 05C85 © 2024, CC BY.

关键词： Undirected graphs

来源：评论

学校读者我要写书评

暂无评论

Deep Residual CNN for Multi-Class Chest Infection Diagnosis

arXiv

引用

arXiv 2023年

作者： Kwon, Ryan Donghan Lim, Dohyun Lee, Yoonha Lee, Seung Won Department of Science and Informatics Hana Academy Seoul Eunpyeong-gu Seoul03305 Korea Republic of Department of Mathematics and Computer Science Korea Science Academy of KAIST Busanjin-gu Busan47162 Korea Republic of Department of Precision Medicine School of Medicine Sungkyunkwan University Jangan-gu Gyeonggi-do Suwon-si16419 Korea Republic of

The advent of deep learning has significantly propelled the capabilities of automated medical image diagnosis, providing valuable tools and resources in the realm of healthcare and medical diagnostics. This research delves into the development and evaluation of a Deep Residual Convolutional Neural Network (CNN) for the multi-class diagnosis of chest infections, utilizing chest X-ray images. The implemented model, trained and validated on a dataset amalgamated from diverse sources, demonstrated a robust overall accuracy of 93%. However, nuanced disparities in performance across different classes, particularly Fibrosis, underscored the complexity and challenges inherent in automated medical image diagnosis. The insights derived pave the way for future research, focusing on enhancing the model’s proficiency in classifying conditions that present more subtle and nuanced visual features in the images, as well as optimizing and refining the model architecture and training process. This paper provides a comprehensive exploration into the development, implementation, and evaluation of the model, offering insights and directions for future research and development in the field. © 2023, CC BY-NC-ND.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Parking Occupancy Prediction and Traffic Assignment in a University Environment 10

Parking Occupancy Prediction and Traffic Assignment in a Uni...

引用

10th International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2022

作者： Farag, Mohamed Hilal, Amr El-Tawab, Samy Arab Academy for Science Technology & Maritime Transport Alexandria Egypt Virginia Tech Transportation Institute Virginia Tech BlacksburgVA United States Informatics Lab Virginia Tech BlacksburgVA United States Alexandria University Department of Computer and Systems Engineering Egypt James Madison University College of Integrated Science & Engineering HarrisonburgVA United States

ISBN: (纸本)9781665464635

The fourth industrial revolution has given rise to large-scale data-driven models like smart cities and Intelligent transportation. Within these models, applications like smart parking have been growing rapidly in research and industry. However, different scenarios and environments (e.g., shopping areas, residential places, and business complexes) can require special handling due to the various factors impacting people's schedules and behavior. In this paper, we provide an initial investigation of traffic assignment based on parking prediction for a mid-size university environment where parking is concentrated in three parking garages around the campus. Our initial investigation includes results for parking prediction using a statistical method and plans for an augmenting study using variations of Neural Networks. On top of the parking prediction layer, we propose an application layer that directs and fuses the model predictions to produce the parking options provided to the application user. The presented investigation can help the university administration in their consideration of building additional garages. © 2022 IEEE.

关键词： Forecasting

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：