检索结果-内蒙古大学图书馆

Rendering discrete participating media using geometrical optics approximation

Computational Visual Media 2022年第3期8卷 425-444页

作者： Jie Guo Bingyang Hu Yanjun Chen Yuanqi Li Yanwen Guo Ling-Qi Yan State Key Lab for Novel Software Technology Nanjing UniversityNanjing 210023China Department of Computer Science UC Santa BarbaraUSA

We consider the scattering of light in participating media composed of sparsely and randomly distributed discrete *** particle size is expected to range from the scale of the wavelength to several orders of magnitude greater,resulting in an appearance with distinct graininess as opposed to the smooth appearance of continuous *** fundamental issue in the physically-based synthesis of such appearance is to determine the necessary optical properties in every local *** these properties vary spatially,we resort to geometrical optics approximation(GOA),a highly efficient alternative to rigorous Lorenz–Mie theory,to quantitatively represent the scattering of a single *** enables us to quickly compute bulk optical properties for any particle size *** then use a practical Monte Carlo rendering solution to solve energy transfer in the discrete participating *** proposed framework is the first to simulate a wide range of discrete participating media with different levels of graininess,converging to the continuous media case as the particle concentration increases.

关键词： light scattering geometrical optics approximation(GOA) discrete participating media volume rendering

来源：评论

学校读者我要写书评

暂无评论

AutoCov22: A Customized Deep Learning Framework for COVID-19 Detection

引用

SN computer science 2023年第5期4卷 659页

作者： Bhowmik, Biswajit Varna, Shrinidhi Kumar, Adarsh Kumar, Rahul Ishwarchandra Vidyasagar AIT Lab BRICS Lab Department of Computer Science and Engineering National Institute of Technology Karnataka Surathkal 575025 India

The novel coronavirus disease 2019 (COVID-19) spill has spread rapidly and appeared as a pandemic affecting global public health. Due to the severe challenges faced with the increase of suspected cases, more testing methods are explored. These methods, however, have several disadvantages, such as test complexity and associated problems—sensitivity, reproducibility, and specificity. Hence, many of them need help to achieve satisfactory performance. Motivated by these shortcomings, this work proposes a custom deep neural network framework named “AutoCov22” that detects COVID-19 by exploiting medical images—chest X-ray and CT-scan. First, multiple neural networks extract deep features from the input medical images, including popularly used VGG16, ResNet50, DenseNet121, and InceptionResNetV2. Then, the extracted features are fed to different machine-learning techniques to identify COVID-19 cases. One objective of this work is to quicken COVID-19 detection. Another goal is to reduce the number of falsely detected cases by a significant margin. Comprehensive simulation results achieve a classification accuracy of 99.74%, a precision of 99.69%, and a recall of 98.80% on exercising chest X-ray images. Extended experiment results in accuracy, precision, and recall up to 87.18%, 84.98%, and 85.66%, respectively, in processing CT-scan images. Thus, the AutoCov22 approach demonstrates a promising and plausible best solution over several methods in the state-of-the-art. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： COVID-19 CT-scan images Deep neural networks False prediction reduction Test accuracy Transfer learning techniques X-ray images

来源：评论

学校读者我要写书评

暂无评论

Detection and Localization of Channel-Short Faults in Regular On-Chip Interconnection Networks

引用

SN computer science 2023年第5期4卷 1-18页

作者： Bhowmik, Biswajit Swami Vivekananda NoC Lab BRICS Lab Department of Computer Science and Engineering National Institute of Technology Karnataka Surathkal 575025 India

With the rapid developments in VLSI technology, the communication channels in networks-on-chip (NoCs) can place many wires for sustaining high-performance requirements over the communication bottleneck in multicore, multiprocessor systems-on-chip (MPSoCs). Consequently, NoC channels, due to increased wire density, are exposed to different logic level faults, e.g., short resulting in reliability and yield issues in NoC-based systems. These faults can appear at any stage of the lifetime of a chip. While existing in an NoC communication architecture, the channel-short faults bring the system into various failures that surprisingly cause a significant deviation from its expected performance. In this work, an online, distributed test solution is presented that detects and diagnoses intra-channel and inter-channel short faults and analyzes the effect of these faults on various performance metrics. Fault simulations ensure up to 100% coverage metrics. Network simulation shows insight into the impact of channel shorts in NoC performances. It is observed that the amount of test time is reduced to 10 × concerning a set of prior works while employed for a group of NoCs. It is also seen on these NoCs that average packet latency is improved by 15.14–46.79% while energy consumption is reduced by 13.68–39.13% by the current solution than the set of existing solutions. Moreover, the proposed solution scales to all NoCs irrespective of size, topology, and channel width at an acceptable test cost. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Defect-based testing Diagonal node selection Fault modelling and evaluation Intra-channel and inter-channel manufacturing shorts Model scalability Networks on a chip Performance analysis and measurement System level failures

来源：评论

学校读者我要写书评

暂无评论

Compound Jamming Detection Technique under High Similarity Background

Compound Jamming Detection Technique under High Similarity B...

引用

2024 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications technology, IAICT 2024

作者： Li, Zhaobin Zhang, Jiaxiang Liu, Quanhua Zhao, Sanyuan Radar Research Lab Department of Computer Science & Technology Beijing Institute of Technology Beijing China Radar Research Lab School of Information and Electronics Beijing Institute of Technology Beijing Institute of Technology Beijing China Beijing Institute of Technology Ministry of Education Beijing Institute of Technology Chongqing Innovation Center Chongqing Beijing China Department of Computer Science & Technology Beijing Institute of Technology Beijing China

ISBN: (纸本)9798350353464

To address the issue of easy misidentification due to the similarity between suppression jamming and background noise on the time-frequency graph in radar jamming detection, this paper proposes a signal amplitude detection method based on empirically preset thresholds to distinguish them. To address the problem of misidentifying dense false targets and target signals caused by their identical mathematical models and similar features in the time-frequency graph—both appearing as linear shapes with the same tilt angle—we distinguish them by utilizing the evenly spaced distribution pattern of the dense false targets. The simulation results demonstrate that in the compound scenario of suppression jamming and dense false targets jamming, the YOLOv8 model achieves an F1-score of 95.7% for jamming detection. By integrating the outlier rejection module proposed in this paper, this performance can be enhanced to 98.6%. ©2024 IEEE.

关键词： Jamming

来源：评论

学校读者我要写书评

暂无评论

LBA: Matching Theory Based Latency-Sensitive Binary Offloading in IoT-Fog Networks 16

LBA: Matching Theory Based Latency-Sensitive Binary Offloadi...

引用

16th International Conference on COMmunication Systems and NETworkS, COMSNETS 2024

作者： Soni, Priyanka Deshlahre, Omkar Chand Satpathy, Anurag Addya, Sourav Kanti National Institute of Technology Karnataka Cloud and Smart System Services Lab Department of Computer Science and Engineering India Missouri University of Science and Technology RollaMO United States

ISBN: (纸本)9798350383119

The Internet of Things (IoT) is growing more popular with applications like healthcare services, traffic monitoring, video streaming, smart homes, etc. These applications produce an enormous amount of data, so a realistic option in this instance is to offload computational tasks to their proximity fog nodes (FNs) instead of the remote cloud. However, a negligent offloading strategy may cause anomalous computational traffic load at the FNs, causing congestion that may adversely affect the latency. However, the latency of task flows from IoT devices comprises communications latency at BS and computational latency at FNs. Therefore, designing offloading algorithms to distribute the computational load at FN evenly and efficiently utilize the FN resources is crucial. To solve this problem, we proposed LBA in a fog network with a binary offloading strategy using the matching theory-based approach. We utilize the Analytic Hierarchy Process (AHP) to generate the preference list. Furthermore, the binary offloading technique follows the deferred acceptance algorithm (DAA) to produce a stable assignment, and the complete offloading problem is modeled as a one-to-many matching game. Comprehensive simulations ensure that LBA can accomplish a better-balanced assignment for homogeneous and heterogeneous input concerning all the baseline algorithms. © 2024 IEEE.

关键词： Analytic Hierarchy Process (AHP) Binary offloading Fog computing IoT Load balancing Matching Theory

来源：评论

学校读者我要写书评

暂无评论

VQR: vulnerability analysis in quadratic residues-based authentication protocols

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第6期14卷 7559-7574页

作者： Ghahramani, Meysam HaddadPajouh, Hamed Javidan, Reza Kumari, Saru Computer Engineering and IT Department Shiraz University of Technology Shiraz Iran Cyber Science Lab School of Computer Science University of Guelph 50 Stone Rd E GuelphON Canada Department of Mathematics Ch. Charan Singh University Uttar Pradesh Meerut India

Ensuring security and respect for users’ privacy, especially in electronic health-care systems, is an important task that is achievable by authentication protocols. The security of many protocols is based on public-key cryptography, the breaking of which depends on solving a hard problem, e.g. Quadratic Residue (QR) which is compatible with constrained devices and is based on integer factoring with the Mean time complexity of exp((c+ o(1)) log 1 / 3(n) log 2 / 3(log (n))) for some c2(n)) to calculate user’s session keys and finding confidential user’s data by passive monitoring of the transmitted data over public networks for 2048-bit public keys in 6.9 μs. To prove the efficiency of the proposed attacks, we address 12 vulnerable QR-based authentication protocols and show that the structure of all vulnerable transmitted messages is limited to 6 cases while messages in (n,n-n) are safe from the proposed method. Also, two recent authentication protocols have been analyzed in detail to show the consequences of the proposed vulnerability. © 2023, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Public key cryptography

来源：评论

学校读者我要写书评

暂无评论

INCENTIVE-AWARE FEDERATED LEARNING WITH TRAINING-TIME MODEL REWARDS 12

INCENTIVE-AWARE FEDERATED LEARNING WITH TRAINING-TIME MODEL ...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Wu, Zhaoxuan Amiri, Mohammad Mohammadi Raskar, Ramesh Low, Bryan Kian Hsiang Institute of Data Science National University of Singapore Singapore Department of Computer Science Rensselaer Polytechnic Institute United States Media Lab Massachusetts Institute of Technology United States Department of Computer Science National University of Singapore Singapore

In federated learning (FL), incentivizing contributions of training resources (e.g., data, compute) from potentially competitive clients is crucial. Existing incentive mechanisms often distribute post-training monetary rewards, which suffer from practical challenges of timeliness and feasibility of the rewards. Rewarding the clients after the completion of training may incentivize them to abort the collaboration, and monetizing the contribution is challenging in practice. To address these problems, we propose an incentive-aware algorithm that offers differentiated training-time model rewards for each client at each FL iteration. We theoretically prove that such a local design ensures the global objective of client incentivization. Through theoretical analyses, we further identify the issue of error propagation in model rewards and thus propose a stochastic reference-model recovery strategy to ensure theoretically that all the clients eventually obtain the optimal model in the limit. We perform extensive experiments to demonstrate the superior incentivizing performance of our method compared to existing baselines. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

An Efficient Long Short-Term Memory and Gated Recurrent Unit Based Smart Vessel Trajectory Prediction Using Automatic Identification System Data

引用

computers, Materials & Continua 2024年第10期81卷 1789-1808页

作者： Umar Zaman Junaid Khan Eunkyu Lee Sajjad Hussain Awatef Salim Balobaid Rua Yahya Aburasain Kyungsup Kim Department of Computer Engineering Chungnam National UniversityDaejeon34134Republic of Korea Department of Environmental IT Engineering Chungnam National UniversityDaejeon34134Republic of Korea Autonomous Ship Research Center Samsung Heavy IndustriesDaejeon34051Republic of Korea ICAROS Center(PRISMA Lab) University of Naples Federico IINaples80131Italy Department of Computer Science College of Computer Science and Information TechnologyJazan UniversityJazan45142Saudi Arabia

Maritime transportation,a cornerstone of global trade,faces increasing safety challenges due to growing sea traffic *** study proposes a novel approach to vessel trajectory prediction utilizing Automatic Identification System(AIS)data and advanced deep learning models,including Long Short-Term Memory(LSTM),Gated Recurrent Unit(GRU),Bidirectional LSTM(DBLSTM),Simple Recurrent Neural Network(SimpleRNN),and Kalman *** research implemented rigorous AIS data preprocessing,encompassing record deduplication,noise elimination,stationary simplification,and removal of insignificant *** were trained using key navigational parameters:latitude,longitude,speed,and *** aware processing through trajectory segmentation and topological data analysis(TDA)was employed to capture dynamic *** using a three-month AIS dataset demonstrated significant improvements in prediction *** GRU model exhibited superior performance,achieving training losses of 0.0020(Mean Squared Error,MSE)and 0.0334(Mean Absolute Error,MAE),with validation losses of 0.0708(MSE)and 0.1720(MAE).The LSTM model showed comparable efficacy,with training losses of 0.0011(MSE)and 0.0258(MAE),and validation losses of 0.2290(MSE)and 0.2652(MAE).Both models demonstrated reductions in training and validation losses,measured by MAE,MSE,Average Displacement Error(ADE),and Final Displacement Error(FDE).This research underscores the potential of advanced deep learning models in enhancing maritime safety through more accurate trajectory predictions,contributing significantly to the development of robust,intelligent navigation systems for the maritime industry.

关键词： Trajectory prediction AIS data smart vessel deep learning LSTM GRU

来源：评论

学校读者我要写书评

暂无评论

Fingerprint Matching for Noisy and Distorted Patterns using a Siamese Network with ResNet50 and Multihead Attention

引用

IEEE Access 2025年 13卷 88047-88064页

作者： Sawhney, Tinny Sharma, Ashu Abrol, Pawanesh Lehana, Parveen Kumar Chaahat Yadav, Vaishali Scherer, Rafal Kulakov, Oleksii AlZubi, Ahmad Ali University of Jammu DSP Lab Department of Electronics India University of Jammu Department of Computer Science and IT India Chitkara University Institute of Engineering and Technology Chitkara University Department of Computer Science and Engineering Punjab India Manipal University Jaipur Jaipur India Czestochowa University of Technology Department of Artificial Intelligence Częstochowa Poland National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" Faculty of Informatics and Software Engineering Kyiv Ukraine King Saud University Community College Department of Computer Science Riyadh Saudi Arabia

Dermatoglyphics, the study of unique ridge patterns on fingertips, plays a crucial role in fingerprint-based identification. However, skin conditions such as psoriasis, eczema, and verruca vulgaris can distort these patterns. This distortion creates challenges for fingerprint matching systems in forensic science and biometric authentication. This paper proposes a Siamese network with a modified ResNet50 architecture and multihead attention mechanisms to improve fingerprint matching under such distortions. The Siamese network enables the system to compare fingerprint pairs effectively, even under distortions. The modified ResNet50 architecture captures intricate ridge patterns, while the multihead attention mechanisms focus on critical fingerprint regions. Overall, the proposed approach enhances the system’s ability to learn discriminative features. As a result, it can effectively differentiate between matched and unmatched fingerprint pairs, even in the presence of moderate to extreme noise. The proposed system was trained on the Sokoto coventry fingerprint dataset (SOCOFing) and a custom dataset. Experimental results demonstrate high accuracy, with the system achieving 99.47% accuracy under low noise conditions and 90% under moderate noise. However, performance declined to 55.56% under extreme distortions. A comparative analysis highlighted the superiority of the proposed system over widely used minutiae-based methods. The latter exhibited a significant drop in matching scores, falling below 15% for highly distorted fingerprints. These findings underscore the limitations of traditional fingerprint recognition techniques and highlight the effectiveness of proposed approach in handling dermatoglyphic distortions. © 2013 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Light gradient boosting machine with optimized hyperparameters for identification of malicious access in IoT network

引用

Digital Communications and Networks 2023年第1期9卷 125-137页

作者： Debasmita Mishra Bighnaraj Naik Janmenjoy Nayak Alireza Souri Pandit Byomakesha Dash S.Vimal Department of Computer Application Veer Surendra Sai University of TechnologyBurlaSambalpur768018OdishaIndia Department of Computer Science Maharaja Sriram Chandra Bhanja Deo(MSCB)UniversityBaripada757003OdishaIndia Department of Software Engineering HaliçUniversity34394IstanbulTurkey Department of Information Technology Aditya Institute of Technology and Management(AITAM)Tekkali532201Andhra PradeshIndia Data Analytics Lab Department of Artificial Intelligence and Data ScienceRamco Institute of TechnologyNorth Venganallur VillageRajapalayam 626117 Virudhunagar DistrictTamilnaduIndia

In this paper,an advanced and optimized Light Gradient Boosting Machine(LGBM)technique is proposed to identify the intrusive activities in the Internet of Things(IoT)*** followings are the major contributions:i)An optimized LGBM model has been developed for the identification of malicious IoT activities in the IoT network;ii)An efficient evolutionary optimization approach has been adopted for finding the optimal set of hyper-parameters of LGBM for the projected ***,a Genetic Algorithm(GA)with k-way tournament selection and uniform crossover operation is used for efficient exploration of hyper-parameter search space;iii)Finally,the performance of the proposed model is evaluated using state-of-the-art ensemble learning and machine learning-based model to achieve overall generalized performance and *** outcomes reveal that the proposed approach is superior to other considered methods and proves to be a robust approach to intrusion detection in an IoT environment.

关键词： IoT security Ensemble method Light gradient boosting machine Machine learning Intrusion detection

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

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

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：