检索结果-内蒙古大学图书馆

5th International Conference on Data Intelligence and Cognitive Informatics, ICDICI 2024

作者： Dhumpati, Raghu Nainwal, Ankita Rekha, G. Sunitha Patil, Harshal Vekariya, Vipul Taqui, Syed Noeman Bahrain Polytechnic Department of Computer Science and Engineering Isa Town3339 Bahrain Graphic Era Deemed to be University Department of Computer Science & Engineering Uttarakhand Dehradun248 002 India Telangana Hyderabad501 510 India Iilm University School of Computer Science & Engineering Greater Noida201306 India Parul Institute of Engineering and Technology Parul University Department of Computer Science and Engineering Post Limda Gujarat Waghodia391760 India Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Vlsi Microelectronics Tamil Nadu Chennai602105 India

ISBN: (纸本)9798350389609

Private message is an innate feature of digital communication and is inherent in maintaining the privacy and confidentiality of the information exchanged in interpersonal and business-to-business relationships. Due to new and more complex cyber threats, there is a need for efficient processing and stronger protection through encryption more than ever before. Conventional cryptographic approaches are prevalent and commonly applied, but they have issues with providing adequate security and carrying out the task with low computational requirements, which poses a problem for devices with limited processing power such as smartphones and IoT devices. However, problems arise with future messaging platforms that include higher latency, scalability, and the changes of different network conditions, which are important for continuous communication on message platforms. Some of the previous efforts include BLIND, LAPEP, PEGA, and SCNN-Bi-LSTM, which have brought improvements addressing some of these challenges. Our research introduces BowNetCrypt: Elegant Vaulted Messenger Secured by Bowerbird-Driven Finite Neural Messaging with Dual Elliptic Protection to address such hindrances. BowNetCrypt also uses the Finite Element Neural Fusion Network (FE-Net) system for better adaptability and accuracy in the private message. The Bowerbird Optimization (BowBO) algorithm is used to minimize the computational complexity as well as enhance the overall network performance. Furthermore, the Lightweight Dual Elliptic Security Protocol (LwDESP) is employed to ensure security and strong encryption without consuming a lot of resources. This method outperforms existing systems by achieving key performance metrics such as 99.2% encryption strength, a 98.7% reduction in latency, and a 97.9% increase in scalability. Such improvements offer BowNetCrypt as one of the most efficient and dependable techniques for implementing secure private messaging within advanced communication networks. © 2024 IEEE.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

Image Recognition in Low-Light Conditions with Deep Learning Model 2

Image Recognition in Low-Light Conditions with Deep Learning...

引用

2nd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2024

作者： Balakrishnan, R. Rawat, Ruchira Kannan, Sathish Pate, Warish Pati, Harshal Rajkumar, S. SMS College of Arts and Science Department of Computer Science Tamil nadu Coimbatore641010 India Graphic Era Deemed to Be University Department of Computer Science and Engineering Uttarakhand Dehradun248002 India Saveetha School of Engineering Department of VLSI Microelectronics Saveetha Institute of Medical and Technical Sciences Tamil Nadu Chennai602105 India Parul Institute of Engineering and Technology Department of Computer Science and Engineering Parul University Post Limda Gujarat Waghodia391760 India Maharashtra Pune412115 India Institute of Technology Faculty of Manufacturing Hawassa University Department of Mechanical Engineering Ethiopia Ethiopia

ISBN: (纸本)9798350327533

Object identification has evolved as a critical application in computer vision during the past decade, experiencing intensive research, and broad practical implementations. While image processing normally works well in regular lighting circumstances, issues develop in low-light scenarios where images become chaotic and dark, causing difficulty for later computer vision tasks. However, conventional object detection algorithms frequently fail in low-light situations. This study investigates obj ect identification in extremely low-light circumstances in detail, outlining critical design concepts required for a robust low-light detection system. Colour distortion, loss of information, and blurriness are common in Low-Light Images (LLI) acquired in poor lighting conditions, reducing the accuracy of obj ect recognition tasks. We use low-light object identification methods like Single Shot Multibox Detector (SSD) and Faster Region-based Convolutional Neural Network (R-CNN), together with image enhancement techniques like Multi-Branch Low-Light Enhancement Network (MBLLEN) and RetinexNet, to improve the precision of obj ect detection in LLI. On the publicly accessible ExDark dataset, the suggested method is trained and evaluated. The experimental findings show that combining MBLLEN image enhancement and SSD object identification resulted in greater precision, recall, and mean Average Precision (mAP) values of 0.77, 0.68, and 0.587, respectively. This study helps to understand and enhance object identification capacity in challenging low-light environments. © 2024 IEEE.

关键词： Deep Learning ExDARK dataset Image enhancement Low Light Object Recognition

来源：评论

学校读者我要写书评

暂无评论

Security Analysis of IoT Networks using Diffusion Kernel Spike Attention Network with Atomic Optimization based Intrusion Detection System 5

Security Analysis of IoT Networks using Diffusion Kernel Spi...

引用

5th International Conference on Data Intelligence and Cognitive Informatics, ICDICI 2024

作者： Sharma, Manju Sharma, Monika Lavanya Sharma, Pramod Gurpur, Shashikala Taqui, Syed Noeman University Institute of Engineering and Technology Maharshi Dayanand University Department of Computer Science and Engineering Haryana Rohtak124001 India The Technological Institute of Textile and Sciences Department of Computer Engineering Haryana Bhiwani127021 India Kalasalingam Academy of Research and Education Department of Computer Science and Engineering Tamil Nadu Viruthunagar626126 India University of Technology School of Engineering & Technology Rajasthan Jaipur303903 India Viman Nagar Maharashtra Pune411014 India Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Vlsi Microelectronics Tamil Nadu Chennai602105 India

ISBN: (纸本)9798350389609

Recently, the most prominent study areas are Internet of Things (IoT) and its applications. The attributes of the Internet of Things (IoT) facilitate its practical uses in real life, but also leave it vulnerable to online attacks. Numerous IoT nodes gather, store, and handle personal data, making them a prime target for hackers. The main objective of a network intrusion detection system is to find network intrusions, which is essential to guaranteeing the Internet of Things' security. Numerous investigators have tackled this issue and have put out numerous intrusion detection methods. Cyberattack subcategories are hard to identify and performance improvement is a challenge for the existing system. This research project introduces a new method for security analysis in Internet of Things networks by creating the Diffusion Kernel Spike Attention Network with Atomic Optimization (DifK-SAN+AO). This research employs a Diffusion Kernel Spike Attention Network to counteract cyberthreats and provides an overall threat analysis of the Internet of Things. The classification of threat and normal patterns on an IoT network is the main topic of this research study. For every form of attack, the proposed model shows promising findings. We use empirically based, simulation-generated datasets (ToN-IoT, BoT-IoT, and CIC-IoT-23 datasets) to detect attacks directed on the IoT ecosystem. In order to perform a security analysis, the results can be compared using the security metrics that are calculated, including the accuracy, precision, recall, and F1-score factors. The experimental results are more reliable than current algorithms and show 99.2% accuracy in detecting a variety of threats. As it would be observed from the results of the simulations, this demonstrates how efficient our method is in assessing the impacts of the attacks in IoT networks and maintaining good performance values. © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Crowd counting in dense UAV images based on multiscale feature fusion 4

Crowd counting in dense UAV images based on multiscale featu...

引用

2023 4th International Conference on computer engineering and Intelligent Control, ICCEIC 2023

作者： Tan, Li Jiang, Xujie Liu, Zikang Zhang, Hongtao Shang, Ziliang Liu, Yuzhao Chongqing Institute of Microelectronics Industry Technology Beijing Technology and Business University School of Computer Science and Engineering Beijing China University of Electronic Science and Technology of China Chongqing China Beijing Technology and Business University School of Computer Science and Engineering Beijing China

ISBN: (纸本)9798350308877

To achieve breakthrough performance in the task of dense crowd counting for unmanned aerial vehicles (UAVs) in complex environments, this paper introduces a multi-scale feature fusion crowd counting network model called MSFFCCNet. This model comprises two main components: a multi-scale feature extraction module and a feature fusion module. The former introduces diversity from multiple scales, where different scales focus on different information and generate scale-related features. The latter is responsible for fusing three features. Experimental results demonstrate that the proposed MSFFCCNet outperforms existing crowd counting network models, exhibiting higher accuracy and robustness. © 2023 IEEE.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

Unitless Unrestricted Markov-Consistent SCM Generation: Better Benchmark Datasets for Causal Discovery

arXiv

引用

arXiv 2025年

作者： Herman, Rebecca J. Wahl, Jonas Ninad, Urmi Runge, Jakob Dresden Leipzig Germany Institute of Data Science Jena Germany Berlin Germany Institute of Computer Engineering and Microelectronics Germany

Causal discovery aims to extract qualitative causal knowledge in the form of causal graphs from data. Because causal ground truth is rarely known in the real world, simulated data plays a vital role in evaluating the performance of the various causal discovery algorithms proposed in the literature. But recent work highlighted certain artifacts of commonly used data generation techniques for a standard class of structural causal models (SCM) that may be nonphysical, including var- and R2-sortability, where the variables’ variance and coefficients of determination (R2) after regressing on all other variables, respectively, increase along the causal order. Some causal methods exploit such artifacts, leading to unrealistic expectations for their performance on real-world data. Some modifications have been proposed to remove these artifacts;notably, the internally-standardized structural causal model (iSCM) avoids varsortability and largely alleviates R2-sortability on sparse causal graphs, but exhibits a reversed R2-sortability pattern for denser graphs not featured in their work. We analyze which sortability patterns we expect to see in real data, and propose a method for drawing coefficients that we argue more effectively samples the space of SCMs. Finally, we propose a novel extension of our SCM generation method to the time series setting. © 2025, CC BY.

关键词： Data assimilation

来源：评论

学校读者我要写书评

暂无评论

Improved Dead-Beat Control for Single-Phase LC-Coupling Hybrid Active Power Filter (LC-HAPF) 32

Improved Dead-Beat Control for Single-Phase LC-Coupling Hybr...

引用

32nd IEEE International Symposium on Industrial Electronics, ISIE 2023

作者： Chan, Pak-Ian Sou, Wai-Kit Lam, Chi-Seng State Key Laboratory of Analog and Mixed-Signal VLSI University of Macau China Institute of Microelectronics University of Macau China University of Macau Department of Electrical and Computer Engineering China

ISBN: (纸本)9798350399714

In this paper, an improved dead-beat controller (DBC) for a single-phase LC-coupling hybrid active power filter (LC-HAPF) is proposed to achieve both low steady-state error and fast transient response. The conventional DBC has an inherent discretized and predictive error, which can increase the steady-state error. Thus, a signal correction technique is used to reduce the steady-state error. Moreover, the improved DBC is implemented with a coupling capacitor voltage observer to replace the voltage sensor, thus reducing the system cost. After that, the stability analysis and parameter design of the proposed improved DBC for the LC-HAPF are given in detail. Subsequently, to verify the effectiveness of the improved DBC, the simulation results are provided in comparison to the conventional DBC. © 2023 IEEE.

关键词： Power quality

来源：评论

学校读者我要写书评

暂无评论

In-situ fabrication of on-chip 1T'-MoTe_(2)/Ge Schottky junction photodetector for self-powered broadband infrared imaging and position sensing

引用

Nano Research 2024年第6期17卷 5587-5594页

作者： Menglei Zhu Kunxuan Liu Di Wu Yunrui Jiang Xue Li Pei Lin Zhifeng Shi Xinjian Li Ran Ding Yalun Tang Xuechao Yu Longhui Zeng School of Physics and Microelectronics and Key Laboratory of Material PhysicsZhengzhou UniversityZhengzhou 450052China Department of Electrical and Computer Engineering University of California San DiegoLa JollaCA 92093USA State Key Laboratory of Integrated Optoelectronics College of Electronic Science and EngineeringJilin UniversityChangchun 130012China Key Laboratory of Multifunctional Nanomaterials and Smart Systems Suzhou Institute of Nano-Tech and Nano-BionicsChinese Academy of SciencesSuzhou 215123China

High-sensitivity room-temperature multi-dimensional infrared(IR)detection is crucial for military and civilian ***,the gapless electronic structures and unique optoelectrical properties have made the two-dimensional(2D)topological semimetals promising candidates for the realization of multifunctional optoelectronic ***,we demonstrated the in-situ construction of high-performance 1T’-MoTe_(2)/Ge Schottky junction device by inserting an ultrathin AlOx passivation *** good detection performance with an ultra-broadband detection wavelength range of up to 10.6 micron,an ultrafast response time of~160 ns,and a large specific detectivity of over 109 Jones in mid-infrared(MIR)range surpasses that of most 2D materials-based IR sensors,approaching the performance of commercial IR *** on-chip integrated device arrays with 64 functional detectors feature high-resolution imaging capability at room *** these outstanding detection features have enabled the demonstration of position-sensitive detection *** demonstrates an exceptional position sensitivity of 14.9 mV/mm,an outstanding nonlinearity of 6.44%,and commendable trajectory tracking and optoelectronic demodulation *** study not only offers a promising route towards room-temperature MIR optoelectronic applications,but also demonstrates a great potential for application in optical sensing systems.

关键词： MoTe_(2) broadband photodetection Schottky junction imaging position sensitive detector

来源：评论

学校读者我要写书评

暂无评论

Interface Ion-Driven, Highly Stable Synaptic Memristor for Neuromorphic Applications

引用

ACS Applied Electronic Materials 2023年第4期5卷 2439-2446页

作者： Gawai, Umesh Wu, Chien-Hung Kumar, Dayanand Chang, Kow-Ming UST-IPPP College of Electrical and Computer Engineering National Yang Ming Chiao Tung University Hsinchu30010 Taiwan Department of Optoelectronics and Materials Engineering Chung Hua University Hsinchu30012 Taiwan Institute of Microelectronics Department of Electrical Engineering National Cheng-Kung University Tainan701 Taiwan Institute of Electronics Engineering National Yang Ming Chiao Tung University Hsinchu30010 Taiwan

Because of the continual advancements in the artificial intelligence technology, its practical applications (e.g., in computer vision, health care, and pattern recognition) are expanding. Various architectures integrating memory cells and transistors have been used to demonstrate artificial synaptic arrays. However, designing memory cells with superior analogue switching is a major challenge in the development of neuromorphic systems. We evaluated a TiN/AlN/Cu/AlN/Pt memristive synapse for neuromorphic computing. The synaptic device exhibited multilevel characteristics with varying reset stop voltages ranging from −2.7 to −2.2 V. The device also exhibited highly stable reparative 200 potentiation and depression cycles with high nonlinearity values of approximately 2.39 for potentiation and −2.19 for depression. The device further exhibited highly stable DC endurance over 1000 cycles, AC pulse endurance (1 M), and stable retention (104 s) at 100 °C without any degradation. The experimental potentiation and depression data were used to train a Hopfield neural network to effectively identify input images with a resolution of 28 × 28 pixels, representing 784 synapses used in the simulation process. The training resulted in an accuracy of >91% in 28 epochs. Therefore, the newly designed device exhibits promising features, such as high linearity and accuracy, and is thus highly suitable for neuromorphic computing purposes. © 2023 American Chemical Society.

关键词： Memory architecture

来源：评论

学校读者我要写书评

暂无评论

Personalized Health Monitoring Platform with Privacy Preservation in IoT Cloud Networks using Deep Learning 8

Personalized Health Monitoring Platform with Privacy Preserv...

引用

8th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud), I-SMAC 2024

作者： Bhat, Devidas Kansal, Vipashi Kumar, B.N. Mohan Patil, Harshal Vekariya, Daxa Kannan, Sathish Department of Information Science and Engineering Karkala Udupi District Karnataka Nitte India Graphic Era Deemed to be University Department of Computer Science & Engineering Uttarakhand Dehradun248 002 India R R Institute of Technology Department of Electronics and Communication Engineering Karnataka Bangalore560090 India Iilm University School of Computer Science & Engineering Greater Noida201306 India Parul Institute of Engineering and Technology Parul University Department of Computer Science & Engineering Post Limda Gujarat 391760 India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Vlsi Microelectronics Tamil Nadu Chennai602105 India

ISBN: (纸本)9798350376425

Deep learning-based models have demonstrated the potential for personalized health monitoring, creating revolutionary opportunities in healthcare delivery and patient outcomes. In the present study, different deep learning algorithms like LSTM, GRU, CNN and Autoencoders are compared on health data. There are various performance measures for each model such as accuracy, precision, recall and F1-score to see how well it can deal with different types of health data. As a result, LSTM achieved remarkable performance metrics including an accuracy of 0.92, a precision of 0.88, recall of 0.94 and F1 score of 0.91 although it required long training time of about 12 hours. GRU had an accuracy value of 0.89, precision was at 0.85;recall stood at 0.91 and f-1 score was at 0.88 while still having a shorter training time which was ten hours unlike CNN that performed well in image processing with an accuracy level of o'95% and preciseness that amounted to o'92%. Besides this CNN took longer to train since the time was twenty-four hours but not like Autoencoders whose anomaly detection efficiency resulted in low reconstruction errors amounting to only five (5%) percent or less even though the training time is just eight (8) hours with relatively lower accuracies and precisions reaching eighty-seven (87%) percent or eighty-two (82%) percent respectively These findings highlight the multifaceted approach and promise that deep learning models bring to transforming individualized health monitoring, allowing accurate extraction of actionable information from personal biometrics and enabling proactive healthcare. These models, which employ advanced computational tools, offer exciting opportunities for improving the speed and precision of diagnoses, refining treatment strategies, and advancing personalized medicine globally. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficient Resource Allocation for Smart Grids using Deep Reinforcement Learning in IoT Cloud Systems 8

Energy-Efficient Resource Allocation for Smart Grids using D...

引用

8th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud), I-SMAC 2024

作者： Patil, Harshal Kansal, Vipashi Balasubramani, R. Manohar, Talari Vekariya, Daxa Kannan, Sathish Iilm University School of Computer Science & Engineering Uttar Pradesh Greater Noida201 306 India Graphic Era Deemed to be University Department of Computer Science & Engineering Uttarakhand Dehradun248 002 India Department of Information Science and Engineering Karnataka Nitte574110 India Anantha Lakshmi Institute of Technology and Sciences Department of Electrical and Electronics Engineering Andhra Pradesh Ananthapuramu515721 India Parul Institute of Engineering and Technology Parul University Department of Computer Science & Engineering Post Limda Gujarat 391760 India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Vlsi Microelectronics Tamil Nadu Chennai602105 India

ISBN: (纸本)9798350376425

This study describes the use of deep reinforcement learning (DRL) models to fine-tune resource allocation for smart grids in order to reduce energy waste and operational costs. The research experiment evaluates four different DRL algorithms, which include SAC, DDPG, PPO, and A3C. It demonstrates performance through various evaluation metrics, which include impeccable energy efficiency, perfect resource utilization, load balancing ratio, average response time, and computational overhead. SAC is proven to achieve 92.5% energy efficiency, 88.7% resource utilization, 90% load balancing ratio, and response time of 120ms. The computational overhead assessed supplements to 15 GFLOPS. DDPG showcased 90.1% energy efficiency and 85.4% resource utilization, 88.9% load balancing, a response time of 110 ms, and a computational overhead of 14.8 GFLOPS. On the other hand, PPO and A3C also proved to work well and have a promising future to be implemented in a dynamic grid. This analysis demonstrates that practical applications from DRL, including real-time decision-making and adaptive control approaches, can prove vital while assigning and working on energy tasks and renewable conduct. These models prove important and beneficial in enhancing the operational efficiency of the smart grid, understanding and learning continuously from real-time data. It will be a vital concept and idea that would be helpful in ensuring that smart grids work perfectly to help in a reduction in carbon footprints and effective energy resource management to ensure end-use service provider satisfactions. © 2024 IEEE.

关键词： Deep reinforcement learning

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：