检索结果-内蒙古大学图书馆

International Conference on Knowledge and Smart technology (KST)

作者： S Alagesan R Murugan Suganya Devi K Ramanujam E Department of Computer Science and Engineering National Institute of Technology Silchar Silchar Assam India Department of Electronics and Communication Engineering National Institute of Technology Silchar Silchar Assam India

ISBN: (数字)9798331520403

ISBN: (纸本)9798331520410

An epileptic seizure is characterized by spontaneous, recurrent abnormal activity in the central nervous system, often presenting clinically as unusual motor, sensory, or psychomotor experiences. Epileptic seizure is a common chronic neurological disorder that can affect people of any age, race, social class, or location. This paper presents an approach that utilizes Continuous Wavelet Transform (CWT) to convert raw EEG signals into time-frequency images, which are then processed by the VGG-19 deep network model for feature extraction and the classification of epileptic seizures. Electroencephalograms (EEG) time-frequency images from the publicly accessible University of Bonn dataset were used to test the proposed model. The model performed exceptionally well with the state-of-the-art regarding accuracy, sensitivity, specificity, and F1-score. This approach represents a significant advancement in EEG-based epileptic seizure disorder classification, promising more accurate and reliable diagnostic tools.

关键词： Deep learning Neurological diseases Time-frequency analysis Continuous wavelet transforms Accuracy Sensitivity Brain modeling Motors Electroencephalography Reliability

来源：评论

学校读者我要写书评

暂无评论

A Framework for Deepfake Detection using Convolutional Neural Network and Deep Features

引用

Procedia computer Science 2025年 258卷 3640-3648页

作者： Soundarya B C Gururaj H L Naveen Kumar C M Department of Information Technology Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India Department of Computer Science and Business System Malnad College of Engineering Hassan India

With the advancement of Artificial Intelligence, facial recognition has become a crucial biometric feature. Deepfake technology leverages AI and can create hyper-realistic digitally manipulated videos of people appearing to say or do things that never occurred. The emergence of Generative Adversarial Networks (GANs) has further enabled the creation of fake visual content with astonishing realism. This technology has diverse applications, such as in the film industry, where it allows for video recreation without reshooting, creating awareness videos, restoring the voices of those who have lost them, and updating movie scenes at low cost. However, this rapid advancement also presents significant challenges. The proliferation of synthetic images raises severe concerns about their societal impact, particularly in terms of potential misuse for harassment and blackmail. Therefore, developing robust deepfake detection models is imperative. This study evaluates the performance of a proposed ResNet34 model in deepfake detection. We utilize the FaceForensics++ dataset to train and assess the model, incorporating images generated by four popular deepfake techniques. Our experimental results demonstrate that integrating linear ternary patterns (LTP) and edge detection-based features with the modified ResNet34 model achieves superior performance, attaining 97.5% accuracy and surpassing other approaches.

关键词： Deepfake Artificial intelligence Deep learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing Honeybee Hive Health Monitoring: Vision Transformer-Based Non-Invasive Classification

Enhancing Honeybee Hive Health Monitoring: Vision Transforme...

引用

Ambient Intelligence in Health Care (ICAIHC), International Conference on

作者： Durga Naeem Ahmad Department Of Computer Applications National Institute of Technology Raipur India

ISBN: (数字)9798331530259

ISBN: (纸本)9798331530266

Monitoring honeybee-hive health is crucial to ensure the continuing existence of honeybee colonies, which are essential to world ecosystems and agriculture. Classical approaches to diagnosing issues in hives, such as the presence of hive beetles or a Varroa mite infestation, can be laborious and need specific expertise. In this study, we present an automated technique for classifying honeybee-hive health using Vision Transformer (ViT) models. Our dataset includes classes that indicate a range of factors, including hive beetles, ant problems, healthy bees, Varroa mite presence, hive robbery, and missing queens. We selected the ViT model for its ability to capture complex patterns in images, making it particularly suitable for the delicate and challenging task of classifying honeybee hive health. The proposed approach yielded favorable results and demonstrated higher accuracy on the testing dataset, indicating its potential to aid in the early detection and management of issues related to the health of the honeybee hive. This study makes precision apiculture better by providing a scalable, effective, and efficient way to rate the health of honeybee hives. It also has implications for improving hive maintenance and honeybee conservation campaigns.

关键词： Deep learning computer vision Accuracy Biological system modeling Soft sensors Transformers Robustness Environmental management Monitoring Testing

来源：评论

学校读者我要写书评

暂无评论

Leveraging Nature-Inspired Algorithms for Feature Selection in Sentiment Analysis: An Evaluation of Particle Swarm Optimization Effectiveness

Leveraging Nature-Inspired Algorithms for Feature Selection ...

引用

2025 International Conference on Visual Analytics and Data Visualization, ICVADV 2025

作者： Dua, Dwiz Singh, Avtar National Institute of Technology Department of Computer Science Jalandhar India

ISBN: (纸本)9798331521394

The Sentiment Analysis deals with algorithms that recognize and classify views expressed in textual data. It is an important piece of natural language processing (NLP). This research emphasizes how SA serves as one of the essential powerful tools by which businesses are enabled to understand customers' attitudes, preferences and behaviors. Analysis of unstructured data from social media reviews and feedback can lead to insights from which businesses make important decisions, drive customer satisfaction and boost brand reputations. There is much importance on sentiment analysis as it converts raw textual data into a meaningful conversion into text. It aids organizations in understanding customer opinions about their brands easily in this online generation and watching trends emerge. Reactive provision becomes possible here against customer expectations and possible issues that could arise from such an opinion landscape. This research also investigates the study of models on deep learning and other machine learning approaches for sentiment analysis. The relevant feature selection from much bigger datasets is one of the key challenges. Natureinspired algorithms such as Particle Swarm Optimization (PSO) would prove important in this regard. PSO enhances the overall performance of sentiment analysis models through the iterative traversal of feature subsets by modeling social behavior. By applying PSO for feature selection, researchers can successfully identify the most salient features that affect sentiment classification significantly. This not only reduces the computational cost but also improves accuracy of the model. This paper covers a complete introductory survey on sentiment analysis and emphasizes how particle swarm optimization may revolutionize analytical skills. © 2025 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Brain-Conditional Multimodal Synthesis: A Survey and Taxonomy

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第5期6卷 1080-1099页

作者： Mai, Weijian Zhang, Jian Fang, Pengfei Zhang, Zhijun South China University of Technology School of Automation Science and Engineering Guangzhou510640 China Southeast University School of Computer Science and Engineering Nanjing210096 China South China University of Technology Key Library of Autonomous Systems and Network Control Ministry of Education The School of Automation Science and Engineering Guangzhou510640 China Institute for Super Robotics Huangpu Guangzhou510555 China Nanchang University Jiangxi Thousand Talents Plan Nanchang330031 China Jishou University College of Computer Science and Engineering Jishou416000 China Guangdong Artificial Intelligence and Digital Economy Laboratory Pazhou Lab Guangzhou510335 China Shaanxi University of Technology Shaanxi Provincial Key Laboratory of Industrial Automation School of Mechanical Engineering Hanzhong723001 China Changsha Normal University School of Information Science and Engineering Changsha410100 China Guangdong University of Petrochemical Technology School of Automation Science and Engineering Institute of Artificial Intelligence and Automation Maoming525000 China

In the era of artificial intelligence generated content (AIGC), conditional multimodal synthesis technologies (e.g., text-to-image) are dynamically reshaping the natural content. Brain signals, serving as potential reflections of how the brain interprets external information, exhibit a distinctive one-to-many correspondence with various external modalities. This correspondence makes brain signals emerge as a promising guiding condition for multimodal synthesis (e.g., image, text, and audio), which is crucial for developing practical brain–computer interface systems and unraveling complex mechanisms underlying human perception. This survey comprehensively examines the emerging field of brain-conditional multimodal synthesis, termed AIGC-brain, to delineate the current landscape and future directions. To begin, related neuroimaging datasets and generative models are introduced as the foundation of AIGC-brain decoding and analysis. Next, we present a comprehensive taxonomy according to AIGC-brain methodologies, followed by task-specific representative work and implementation details to facilitate in-depth comparison and analysis. Quality assessments are then introduced for both qualitative and quantitative evaluation. Finally, this survey explores insights gained, outlining current challenges and prospects of AIGC-brain. As a pioneering survey, this article paves the way for future advances in AIGC-brain research. © 2020 IEEE.

关键词： Taxonomies

来源：评论

学校读者我要写书评

暂无评论

Plant Leaf Disease Detection Using Deep Learning Techniques

Plant Leaf Disease Detection Using Deep Learning Techniques

引用

IEEE Students' Conference on Electrical, Electronics and computer Science (SCEECS)

作者： Sakshi V. Sawant Vinod Kimbahune Kapil N. Vhatkar Department of Computer Engineering Dr. D. Y Patil Institute of Technology Pune India

ISBN: (数字)9798331529833

ISBN: (纸本)9798331529840

For the Indian economy to meet the demands of its people and improve the welfare of its farmers, agricultural output is crucial. Crop diseases and pests can significantly lower yield and quality, making increased crop output impossible. Recognizing these problems is critical to precision agriculture operations. Crop leaves and other plant parts can be examined using deep learning to identify diseases and pests. This work looks into deep learning approaches for detecting illnesses and pests in crops and presents a model for automated diagnosis. Farmers must be able to diagnose agricultural illnesses and pests in order to prevent crop harm. An overview of crop disease and pest diagnosis techniques is provided in this abstract. Usually, visual inspection and clues like discolored, wilting, or irregular leaves are used to identify crops. Agronomists and farmers utilize their expertise to recognize pests and illnesses unique to a given crop. But this tactic is arbitrary and liable to be wrong. To address these issues and enhance the effectiveness and precision of identification, cutting-edge technologies have been developed.

关键词： Deep learning Precision agriculture Visualization Crops Pesticides Inspection Convolutional neural networks Reliability Diseases Farming

来源：评论

学校读者我要写书评

暂无评论

Advancing Low-Resource NLP: Contextual Question Answering for Bengali Language Using Llama

Advancing Low-Resource NLP: Contextual Question Answering fo...

引用

International Conference on Electrical, computer and Communication engineering (ECCE)

作者： Koshik Debanath Sagor Aich Azmain Yakin Srizon Department of Computer Science & Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Natural language processing (NLP) has witnessed significant advancements in recent years, particularly in improving question-answering (QA) systems for well-resourced languages such as English. However, the development of such systems for low-resource languages, including Bengali, remains insufficiently explored. This study proposes an approach to developing a Bengali QA system utilizing the Llama-3.2-3B-Instruct model, leveraging transfer learning techniques on a synthetic dataset derived from the SQuAD 2.0 benchmark. The experiments achieved an F1 score of 42.77%, marking a 4.02% improvement over the previous best performance of multilingual BERT (mBERT) variants. These results establish a benchmark against human responses and underscore the potential of transfer learning in advancing QA capabilities for Bengali and similar low-resource languages.

关键词： Adaptation models Large language models Computational modeling Transfer learning LoRa Reinforcement learning Benchmark testing Question answering (information retrieval) Multilingual Synthetic data

来源：评论

学校读者我要写书评

暂无评论

GSAgri: Green and Secure Agriculture through efficient task offloading and scheduling under IoT-enabled energy-harvesting multi-access edge computing framework

引用

Expert Systems with Applications 2025年 284卷

作者： Islam, Akhirul Ghose, Manojit Department of Computer Science and Engineering Indian Institute of Information Technology Guwahati India

Smart and sustainable agriculture is an emerging field that integrates seamlessly with edge computing, the Internet of Things (IoT), and energy harvesting technologies. The IoT devices used in smart agriculture need to offload compute-intensive and energy-hungry tasks of their applications to remote servers due to their resource limitation. This paper presents a cooperative heterogeneous multi-access edge computing (MEC) framework with energy-harvesting IoT devices for efficiently running agricultural applications. We propose a novel strategy called GSAgri, which combines a metaheuristic simulated annealing approach for application task offloading and a heuristic for task scheduling, incorporating a machine learning (ML) model for energy prediction. The proposed strategy significantly outperforms state-of-the-art solutions, reducing average application latency by 19.19%, user device energy consumption by 62.76%, number of application failures by 73.68%, and critical task failures by 13.08%. We further benchmark our strategy using a real-world ML-based disease prediction application for tomato leaves, demonstrating improvements of 18.74% in latency and 80.55% in user device energy consumption over baseline methods. Additionally, we evaluate the GSAgri strategy under real-world network conditions using a publicly available dataset, where it achieves 16.57% lower latency and 70.85% reduction in energy consumption compared to the benchmark strategies. The key reasons behind the performance of our proposed strategy are the cooperative MEC system model, the efficient task offloading and scheduling algorithm, and the predictive energy harvesting model. © 2025 Elsevier Ltd

关键词： Scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

Grid-Interactive Novel Resilient Control of Solar PVWind-Battery Storage Microgrid Under Distorted and Unbalanced Grid Voltages

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Gali, Vijayakumar Jamwal, Prashant Kumar Gupta, Nitin Babu, B. Chitti Ahmadi, Mohammad Jawid Morey, Meghraj Sudhakar University of Hertfordshire School of Physics Engineering and Computer Science Department of Electrical and Electronics Engineering HatfieldAL109AB United Kingdom Nazarbayev University Department of Electrical and Computer Engineering Astana Kazakhstan Malaviya National Institute of Technology Department of Electrical Engineering Jaipur India Indian Institute of Information Technology Design and Manufacturing Department of Electronics and Communication Kancheepuram India

Integrating solar photovoltaic (PV), wind, and battery storage (BS) systems into the grid introduces significant power quality (PQ) challenges. In particular, the intermittent nature of solar PV and wind energy systems (WES), combined with nonlinear loads, can lead to grid instability. As a result, maintaining a reliable and high-quality power supply to consumers becomes a substantial challenge. This article presents two novel solutions to address these challenges: the anti-windup mixed-order generalized integrator (AWMOGI) and the adaptive delay operation period filter (ADOPF). The AWMOGI effectively extracts fundamental components from distorted utility grid voltages, minimizing DC offset and reducing steady-state oscillations. Meanwhile, the ADOPF swiftly captures positive sequence voltage components by introducing a two-sampling-interval delay in the d-q coordinate system. The hierarchical-loop controller parameters are optimized using a customized Harris Hawks Optimization (HHO) algorithm, which enhances error reduction and facilitates rapid establishment of reference values for DC-link voltage loop control. An adaptive dynamic power management scheme also enables the microgrid (MG) to manage surplus power from MG sources, reducing stress on the BS and ensuring DC-link voltage stability. The experimental results demonstrate significant improvements in handling utility grid abnormalities, dynamic loads, and intermittent microgrid source operation. The experimental results show a zero DC-off set and steady-state errors, a fast computational time of 35e-5s, and the THD of grid currents of 3.16%, which complies with the IEEE 519 standards. © 1975-2011 IEEE.

关键词： Adaptive filters

来源：评论

学校读者我要写书评

暂无评论

A Review: Threat Modeling in Cyber Security Through XDR 12th

A Review: Threat Modeling in Cyber Security Through XDR

引用

12th International Conference on Frontiers of Intelligent Computing: Theory and Applications, FICTA 2024

作者： Shinde, Swapnil Kisan Deshmukh, Suruchi Parikshit Sharma, Parth Pradhan, Sagar Kshirsagar, Kundlik B. Gaikwad, Aarti S. Vishwakarma Institute of Information Technology Maharashtra Pune India MIT Art Design and Technology University Maharashtra Pune India Nagpur Campus Pune India Computer Science and Engineering Arya College of Engineering Rajasthan Jaipur India Department of Mechanical Engineering Dr. D. Y. Patil Institute of Technology Maharashtra Pune Pimpri India D. Y. Patil College of Engineering Maharashtra Pune Akurdi India

ISBN: (纸本)9789819601424

Given the sophistication of today’s attacks, cyber security has taken on more importance in the digital world. Conventional security measures are no longer sufficient;instead, a more sophisticated and reactive reaction is needed. In this regard, handling this escalating threat seems to depend on the integration of machine learning (ML) and artificial intelligence (AI). But even with their great efficacy, there remains a growing issue: integrating different data sources and technologies so that the overall security of the system model is maintained. With a special emphasis on Open Extended Detection and Response (Open XDR) technology, this paper provides a thorough analysis of the integration of AI and ML in cyber security. A thorough literature review, which looks at the interactions and functionality of several cyber security components, is the methodology employed. In the context of AI and ML, the paper also examines the function of Active Directory, a directory service for Windows domain networks, and the procedure known as log forwarding, which involves sending log files to a central server for analysis. The study explores the evolution of AI and ML, highlighting their applications in cyber security for sophisticated data processing, threat detection, and pattern identification. It examines how machine learning approaches, such as supervised learning (where the model is trained on labeled data) and unsupervised learning (where the model learns from unlabeled data), support cyber security protocols. The paper emphasizes Open XDR’s importance as a breaking invention that combines information from several sources to provide thorough security analysis. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Cyber attacks

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：