检索结果-内蒙古大学图书馆

Skin Cancer Prediction by Incorporating Bio-inspired Optimization in Deep Neural Network

SN computer Science 2024年第8期5卷 1-20页

作者： Mundada, Monica R. Sowmya, B.J. Supreeth, S. Prabhu, Smitha G. Mahesh, K. Vishwanath, Y. Rohith, S. Department of Computer Science and Engineering M S Ramaiah Institute of Technology Bengaluru 560054 India Department of Artificial Intelligence and Data Science Ramaiah Institute of Technology Bengaluru 560054 India School of Computer Science and Engineering REVA University Bengaluru 560064 India Department of Electronics and Communication Engineering Nitte Meenakshi Institute of Technology Bengaluru India Department of Electrical and Electronics Engineering Sir M Visvesvaraya Institute of Technology Bengaluru India School of Computer Science & Engineering Presidency University Bengaluru 560064 India Department of Electronics & Communication Engineering Nagarjuna College of Engineering & Technology Bengaluru 562164 India

This study offers a unified surgical conception of skin neoplasms using deep learning and genetic optimization techniques to improve classification values and speed of diagnosis. In this case, the architecture of Convolutional Neural Networks such as MobileNet, Xception and Inception is enhanced on the basis of the creative algorithms of Particle Swarm Optimization and Bat Algorithm to adjust the hyperparameters. The Heroku hosted cloud solution permits registered users to upload pictures of skin lesions via the web interface, where the picture is then processed through a Deep Neural Network (DNN) model that classifies the lesion as either melanoma or non-melanoma. The Convolution Neural Network (CNN) models were trained and evaluated on large publicly available datasets of skin lesions like ISIC, HAM10000 and the model performance was evaluated based on the effect of varying optimizers. Among all the architectures, Xception with Bat Algorithm (BA) performed best because of the more complex and deeper network architecture and extreme inception modules, which made it possible to extract and classify more complicated features. As a cloud-based solution, this system provides dermatologists and health practitioners with a more practical and workable solution for the detection of skin cancers. Furthermore, it can be utilized in other fields, which implies that this approach could extend its applications beyond dermatological imaging or imaging of patients at risk for other diseases, thus creating an adaptable and effective tool for clinical diagnosis in many areas of medicine. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Bio-inspired optimization Deep neural network (DNN) Skin cancer prediction

来源：评论

学校读者我要写书评

暂无评论

Application of a three-phase unified power quality conditioner in a microgrid

引用

IET Smart Grid 2024年第6期7卷 872-890页

作者： Wu, Tsung-Hsun Huang, Shu-Syuan Chen, Pei-Yin Department of Computer Science and Information Engineering National Cheng Kung University Tainan Taiwan Department of Power Electronics Industrial Technology Research Institute Tainan Taiwan

In power distribution systems, common issues such as voltage fluctuations, voltage instability, current harmonics, and power imbalances often arise, negatively impacting the stability and power quality (PQ) of the power system. As an advanced power electronics device, the unified power quality conditioner (UPQC) can effectively address these problems. Therefore, the integration of UPQC into power distribution lines can enhance the stability and reliability of the power system, providing higher-quality power supply to meet the needs of both users and suppliers for PQ. The authors utilise the TI TMS320F28377D chip as the control core to construct a simple yet highly efficient UPQC system. Through simulation, the correctness of the voltage and current compensation algorithms has been verified, and the overall system feasibility has been validated using a 100-kVA series-shunt inverter. The experimental results show that the voltage compensation range of the UPQC is 380V ± 10%. During sudden current drops, the imbalance rate can be reduced from 85.5% to 1.32% after controlled compensation. During sudden current rises, the imbalance rate can be reduced from 13.1% to 1.56% after controlled compensation. These findings demonstrate the effectiveness of the UPQC system in improving PQ and system stability in power distribution networks. © 2024 The Author(s). IET Smart Grid published by John Wiley & Sons Ltd on behalf of The Institution of engineering and technology.

关键词： Power quality

来源：评论

学校读者我要写书评

暂无评论

Combining Fuzzy Logic and Reinforcement Learning for Resource Management in Edge Computing 5

Combining Fuzzy Logic and Reinforcement Learning for Resourc...

引用

5th International Conference on electronics and Sustainable Communication Systems, ICESC 2024

作者： Prasath, S. Ram Selvi, M. Krishnan, R. Santhana Nithila, Ezhil E. Malar, P. Stella Rose Sankar, G. Ram Scad College of Engineering and Technology Computer Science and Engineering Cheranmahadevi India Rajarajeswari College of Engineering Computer Science and Engineering Bangalore India Scad College of Engineering and Technology Electronics and Communication Engineering Chranmahadevi India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Electronics and Communication Engineering Chennai Avadi India J.P. College of Engineering Electronics and Communication Engineering Thenkasi India Velammal College of Engineering and Technology Department of Computer Science and Engineering Madurai India

ISBN: (纸本)9798350379945

With the wide spread of Internet of Things (loT) devices, edge computing plays a decisive part in meeting the computational demands of loT applications. Efficient resource allocation in such networks is essential to ensure optimal utilization of resources and meet the quality of service (QoS) necessities of loT applications. This research addresses the challenge of dynamic resource allocation in edge computing networks for loT applications through a novel hybrid AI approach, 'FledegeRL.' By integrating fuzzy logic and reinforcement learning, FledegeRL aims to optimize resource utilization and enhance system performance. The proposed framework is evaluated through simulation analysis and compared with existing algorithms, including Q-Iearning, Fuzzy Logic Control, and Ant Colony Optimization, using metrics such as resource allocation accuracy, average throughput, and network overhead. The results demonstrate that FledegeRL consistently outperforms these algorithms, achieving higher accuracy, greater throughput, and lower network overhead. This highlights the potential of FledegeRL as an effective solution for dynamic resource allocation in edge computing environments. © 2024 IEEE.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Crowd Size Estimation: Smart Gathering Management 2nd

Crowd Size Estimation: Smart Gathering Management

引用

2nd International Conference on Emerging Trends in engineering and technology, EGTET 2022

作者： Swami, Ishita Das, Nimish Sunil Department of Computer Engineering LJ Institute of Engineering and Technology Ahmedabad India Department of Electronics and Communication Engineering LJ Institute of Engineering and Technology LJK University Ahmedabad India

ISBN: (纸本)9789819943616

Linear increase in population which results in overcrowding has become an unavoidable element in any public gathering. Public safety under such condition has become a very vital problem in areas like streets, malls and railway stations during weekends, festive seasons, holidays, concerts, etc., normally or in any pandemic situation. The massive disasters that can occur includes numerous instances of fatality where people gather in form of throng. In present time, surveillance cameras are deployed to maintain peace, security and manage crowd, as surveillance videos for proper analysis of crowd activities is an important issue for communal harmony and security;however, some major limitations in video surveillance system are that includes picture getting blurred, peculiarities among person cannot be identified automatically with respect to surroundings during live video streaming, along with that to save the information a lot of storage spaces is also required and hence it becomes costly to run and maintain. The present study proposes a method that is based on principle of Histogram of Oriented Gradients (HOG) and OpenCV that efficiently keeps in track count of the people in the scene which helps in efficient crowd management. OpenCV-based method used for crowd estimation written in Python used in this study in order to count the number of heads in live streaming and helps in crowd management according to requirement in an economical way. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2024.

关键词： Python

来源：评论

学校读者我要写书评

暂无评论

A Secured Online Voting system using Visual Cryptography and Validation 4

A Secured Online Voting system using Visual Cryptography and...

引用

4th International Conference on Intelligent Technologies, CONIT 2024

作者： Sable, Sayali Ghadage, Tanuja Shelurkar, Arpit Moon, Sunil K. SCTR's Pune Institute of Computer Technology Department of Electronics and Telecommunication Engineering Pune India

ISBN: (纸本)9798350349900

Phishing poses a substantial threat to online voting, demanding robust prevention methods. Visual Cryptography (VC) offers promise, enhancing security and preserving voting system confidentiality. The Voting System Using VC enables secure remote voting, merging two VC shares to create a unique password - one emailed to voters before the election, the other accessible during the election. This combats phishing, ensuring secure ballot casting. Amid rising phishing risks, innovative prevention techniques are explored, including image-based VC verification and biometric finger sensors. The paper emphasizes internet voting's security, privacy, and stakeholder involvement, concluding that integrating VC and advanced anti-phishing measures comprehensively secures online voting systems against evolving threats. Additionally, online voting facilitates convenient voting for individuals from diverse locations and aids disabled individuals in casting votes from any place. This emerging field underscores the importance of addressing major threats to ensure the integrity and accessibility of the electoral process. © 2024 IEEE.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

Search Engine for QnA using Distributed Inverted Index System 3

Search Engine for QnA using Distributed Inverted Index Syste...

引用

3rd International Conference for Innovation in technology, INOCON 2024

作者： Dey, Snehasis Moharana, Bhimasen De, Utpal Chandra Samant, Tapaswini Behera, Trupti Mayee Banerjee, Shobhan College of Engineering Electronics & Telecommunication Engineering Bhubaneswar India Lovely Professional University Department of Computer Science & Engineering Punjab India Kalinga Institute of Industrial Technology School of Computer Applications Bhubaneswar India Kalinga Institute of Industrial Technology School of Electronics Engineering Bhubaneswar India Birla Institute of Technology and Science Department of Engineering Technology Pilani India

ISBN: (纸本)9798350381931

The internet along with its global uses has also evolved as a platform where various people can pose their questions on certain sites and get them answered by others from the same domain. Certain sites like Quora, Reddit, Yahoo Answers, etc. are sites enabling people to enhance their range over the net to find solutions to various issues or answers to various questions. But this leads to redundancy in data & hence increased memory consumption where different users can pose the same question in different ways. The same thing stands for the answers posted by different users for the same question in various ways. In this paper, we have worked on a search engine system for Question & Answers where given a question, we can find similar questions with the same semantic meaning using Elastic Search which uses an inverted index system & looks at various aspects of the formulated problem statement. © 2024 IEEE.

关键词： Search engines

来源：评论

学校读者我要写书评

暂无评论

Three-Port DC-DC Converter with Galvanic Isolation Using Multi-Winding Transformer 3

Three-Port DC-DC Converter with Galvanic Isolation Using Mul...

引用

3rd IEEE International Conference on Power electronics, Intelligent Control and Energy Systems, ICPEICES 2024

作者： Paul, Preenu Jose, Babita R. Shahana, T.K. Abraham, Chikku Mathew, Jimson School of Engineering Cochin University of Science and Technology Division of Electronics Engineering Kerala India Muthoot Institute of Technology and Science Department of Electrical and Electronics Engineering Kerala India Indian Institute of Technology Department of Computer Science and Engineering Patna India

ISBN: (纸本)9798350387193

A three port DC-DC converter that employs push-pull topology with isolation is presented. The converter transfers power from two DC input sources using a multi-winding transformer to the load. The high gain of the converter is achieved using quasi - switched boost (qSB) network, switched capacitor (SC) topology, multi winding transformer turn's ratio and voltage doubler circuit (VDR). The converter operation, steady state analysis deriving output voltage and comparison with similar topologies are presented. MATLAB/ Simulink results justify the converter operation and analysis. The underlined features of proposed converter are (i) high voltage conversion ratio (ii) galvanic isolation (iii) reduced utilization of semiconductor devices and (iv) capability to combine multiple input sources. © 2024 IEEE.

关键词： Capacitor bank

来源：评论

学校读者我要写书评

暂无评论

Rotation-Invariant LiDAR-based Localization Method Based on Feature Pyramid and Vector Neural Network 4

Rotation-Invariant LiDAR-based Localization Method Based on ...

引用

4th International Conference on Neural Networks, Information and Communication engineering, NNICE 2024

作者： Tian, Gengxuan Zhao, Junqiao Tongji University School of Electronics and Information Engineering Department of Computer Science and Technology Shanghai China

ISBN: (纸本)9798350394375

LiDAR-based localization for unmanned vehicle has recently become a highly prominent research area. The LiDAR-based localization task typically involves two crucial tasks: place recognition and pose estimation. However, current research often fails to effectively address issues of place recognition retrieval failures caused by rotations. Some existing methods employ a strategy of coupling the pose estimation task with location recognition, but they often separate these two tasks using two distinct sets of networks to achieve robustness against rotations. Unfortunately, such methods frequently exhibit poor generalization. In this paper, we propose an innovative rotation-invariant LiDAR-based localization method based on feature pyramids and vector neural networks. This approach extracts rotation-invariant global features to accomplish the location recognition task and utilizes rotation-invariant local features for matching corresponding points in point clouds. Employing a coarse-to-fine point matching filtering method to eliminate outliers, and ultimately implementing pose estimation through the RANSAC algorithm, we successfully achieve high-precision and rotation-robust LiDAR-based localization. The experiments indicate that on the Oxford dataset, our method consistently achieves a remarkable 94.7% AR@1% regardless of whether the point cloud undergoes rotation, a performance significantly superior to other methods © 2024 IEEE.

关键词： Rotation

来源：评论

学校读者我要写书评

暂无评论

Review and Explore the Transformative Impact of Artificial Intelligence (AI) in Smart Healthcare Systems

Review and Explore the Transformative Impact of Artificial I...

引用

2024 International Conference on Advances in Computing Research on Science engineering and technology, ACROSET 2024

作者： Nagar, Sneha Patidar, Mukesh Sheikh, Rashid Singh, Sudhanshu Kashiv, Ashutosh Jain, Ankit Namdeo, Shravan Kumar Paranjpe, Pranav Mandloi, Devendra Singh Indore Institute of Science and Technology Department of Electronics and Communication Engineering Indore India Parul Institute of Engineering and Technology Parul University Department of Computer Science & Engineering Vadodara India Acropolis Institute of Technology and Research Department of Computer Science & Engineering Indore India Indian Institute of Technology Indore Department of Electronics and Communication Engineering Indore India

ISBN: (纸本)9798350388800

Because of its high level of complexity and data collection capabilities, the healthcare industry employs artificial intelligence (AI) more than other sectors. Artificial intelligence is becoming an increasingly important part of our daily routines. AI's features include extremely low error rates, constant accessibility, quick analysis, and real-time data dissemination. This paper examines the crucial role of artificial intelligence in intelligent healthcare systems, with a particular emphasis on enhancing healthcare delivery. Artificial intelligence in healthcare is rapidly evolving, with the potential to significantly enhance patient care, diagnosis, treatment, and cost distribution. It also presents issues of transparency, honesty, and availability. This study will look into the benefits, drawbacks, and other aspects of artificial intelligence in intelligent healthcare systems. © 2024 IEEE.

关键词： Artificial Intelligence (AI) Ethical Considerations Explainable AI (XAI) Healthcare Resource Allocation Healthcare Systems Smart Treatment Planning

来源：评论

学校读者我要写书评

暂无评论

Enhanced perceptual wavelet packet features for spontaneous Kannada sentence recognition under uncontrolled conditions

引用

International Journal of Speech technology 2025年第1期28卷 153-174页

作者： Shanthamallappa, Mahadevaswamy Pradeep Kumar, B.P. Department of Electronics and Communication Engineering Vidyavardhaka College of Engineering Mysuru Visvesvaraya Technological University Karnataka India Department of Computer Science and Design Atria Institute Of Technology Bengaluru Visvesvaraya Technological University Karnataka India

A Robust Automatic Speech Recognition (ASR) system is proposed through a hybrid combination of Perceptual Wavelet Packet features, Deep Neural Network-Hidden Markov Model (DNN-HMM) acoustic models, and n-gram language models for recognizing Spontaneously spoken Kannada sentences. Two Experiments are performed on own Kannada speech corpus and the widely recognized TIMIT corpus. First one is the experiment on the clean speech datasets and the second one is on the speech degraded by noise data chosen from Noise-US dataset. The proposed system performance is evaluated through the metric Word Error Rate (WER). A comparative study on the performance analysis is presented. The results of the experimental investigations achieved an average reduction in Word Error Rate (WER) of 1.8% over Mel Frequency Cepstral Coefficients (MFCCs) and 2.1% over Perceptual Linear Prediction (PLP) features. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Hidden Markov models

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：