检索结果-内蒙古大学图书馆

GaitC3I: Robust Cross-Covariate Gait Recognition via Causal Intervention

IEEE Transactions on Circuits and Systems for Video technology 2025年

作者： Wang, Jilong Hou, Saihui Guo, Xianda Huang, Yan Huang, Yongzhen Zhang, Tianzhu Wang, Liang University of Science and Technology of China School of Information Science and Technology Department of Automation Hefei230026 China Beijing Normal University School of Artificial Intelligence Beijing100875 China Watrix Technology Limited Co. Ltd Beijing100088 China Wuhan University School of Computer Science China Institute of Automation Chinese Academy of Sciences New Laboratory of Pattern Recognition Beijing100190 China

Cross-covariate gait recognition aims to analyze a pedestrian's gait to extract an identity representation that is invariant across varying covariates. However, prevailing methods that have achieved good results on controlled in-the-lab datasets often perform poorly on realistic datasets. In this work, we find a significant cause is that the widely used pairwise metric learning paradigm cannot correctly handle the relationship between samples from different covariate conditions. Even worse, it may yield harmful signals that inadvertently mislead models to focus on covariate-related features, particularly when covariate distributions vary across subjects. To address this issue, we propose a Cross-Covariate Causal Intervention (GaitC3I) framework, a unified causality-inspired approach aimed at enhancing the robustness of gait recognition across diverse conditions. Specifically, our method consists of two parts: (i) an effective causal intervention metric learning paradigm based on backdoor adjustment, which strategically mitigates spurious correlations induced by covariates, thus ensuring a more invariant gait representation;and (ii) an annotation-free selection strategy that progressively matches each positive sample with negative samples from similar covariate conditions at various granularities. We demonstrate the effectiveness of our GaitC3I through extensive evaluation on six popular gait datasets-Gait3D, GREW, OUMVLP, CASIA-B, CCPG, and CCGR-achieving substantial improvements. Our method not only outperforms existing state-of-the-art models but also provides a systematic solution to remove the spurious correlations in gait recognition. © 1991-2012 IEEE.

关键词： backdoor adjustment biometrics causal inference debiasing gait recognition Identification of persons

来源：评论

学校读者我要写书评

暂无评论

AI-Powered Personalized Deep Learning Framework for Early Detection of Glioblastoma Using Multi-Modal MRI Data

AI-Powered Personalized Deep Learning Framework for Early De...

引用

Data science, Agents & Artificial Intelligence (ICDSAAI), International Conference on

作者： Sunitha T Veeramalai Sankaradass V.K Manindra Manish Department of Artificial Intelligence and Data Science Saveetha Engineering College Chennai India Dept. of Computer Science and Engineering Chennai Institute of Technology Chennai India Bachelor of Medicine and Bachelor of Science Dhanalakshmi Srinivasan Institute of Medical Sciences and Hospital Perambalur India

ISBN: (数字)9798331537555

ISBN: (纸本)9798331537562

The prognosis of glioblastoma, which stands as the most malignant variant of a brain tumour, is clearly unfavourable due to the late diagnosis and the severe heterogeneity it is known to possess. Standard diagnostic techniques nowadays that employ MRI data quite often have a low-level performance in terms of diagnosing glioblastoma during its early stages owing to the diverse manner in which the tumours may grow in various patients. In light of this background, we present a new deep learning framework suitable for multi-modal MRI data and aimed at improving glioblastoma prognosis at an early stage. The framework utilizes convolutional neural networks (CNNs) to obtain features from T1, T2, and FLAIR MRI images, and combines this with an AI-guided individualized diagnostic system. This model utilizes relevant clinical data such as age and genetic data in a more individualized manner to meet patient’s needs during diagnosis. One more novel feature of our approach is the ensemble learning that combines CNNs with RNNs to detect minor tumour deposits more effectively over time. The proposed framework delivered an AUC of 0.94 on large glioblastoma MRI databases, thus progressing the early-stage diagnosis of the tumour type which was the aim. This advanced personalized method not only allows for superior diagnosis but also improves on the treatment plans that are available for such patients. Further development will concern clinical verification as well as increasing the capacity of the framework.

关键词： Deep learning Accuracy Sensitivity Magnetic resonance imaging Genetics Feature extraction History Prognostics and health management Standards Tumors

来源：评论

学校读者我要写书评

暂无评论

Enhanced Software Defect Prediction using Quantum Hamiltonian Generative Adversarial Network for Improved Software Performance Reliability

Enhanced Software Defect Prediction using Quantum Hamiltonia...

引用

International Conference on Inventive Computation Technologies (ICICT)

作者： Jyoti Chaudhary Vijaya Kumar A V Manikannan K Amol Sapatnekar Amit Barve Ramya Maranan Department of Computer Engineering The Technological Institute of Textile and Sciences Bhiwani Haryana India Department of Computer Science & Engineering Presidency University Bangalore Karnataka India Department of Computer Science and Engineering R.M.K Engineering College Thiruvalluvar Tamil Nadu India Faculty of Law Symbiosis Centre for Advanced Legal Studies and Research Symbiosis Law School Symbiosis International (Deemed University) Pune India Department of Computer Science and Engineering Parul Institute of Technology Faculty of Engineering and Technology Parul University Gujarat India Department of Research and Innovation Saveetha School of Engineering SIMATS Chennai Tamil Nadu India

ISBN: (数字)9798331512248

ISBN: (纸本)9798331512255

The development of software systems with high quality and decreased maintenance costs depends on Software Defect Prediction (SDF). The current dataset qualities combined with feature redundancies make confident fault detection difficult through conventional methods. Current defect prediction techniques display multiple deficits through inaccurate results along with many false alarms and dataset bias. A high-performance quantum-inspired method must be researched in working to create an accurate precision model. The project will create the Enhanced Software Defect Prediction using Quantum Hamiltonian Generative Adversarial Network for Improved Software Performance Reliability (DBO-QH-GAN) model. Several software metrics need to be retrieved from the PROMISE datasets before this process can be initiated. The second step includes Fuzzy K- Top Matcher (FK- TM) functionality that removes undesirable noise factors such as redundant features and missing values as well as outliers. The Botox Optimization (BotO) algorithm selects five essential metrics that encompass cyclomatic complexity as well as Halstead measures and coupling and code churn and historical defect density. These preprocessed set of characteristics are fed to a Quantum Hamiltonian Generative Adversarial Network (QH-GAN) which identifies the defects from probabilistic thresholding and thereafter optimized using Duck Bevy Optimization (DBO) to achieve the optimal possible performance. Accuracy of the model was 99.2%, precision was 99.4%, and recall was 99.1 %. The DBO-QH-GAN model sets new standards for defect prediction by providing an improved enhancement of reliability along with decreased maintenance cost.

关键词： Quantum computing Costs Accuracy Software performance Predictive models Generative adversarial networks Software reliability Maintenance Optimization Standards

来源：评论

学校读者我要写书评

暂无评论

Combating Toxic Language: A Review of LLM-Based Strategies for Software Engineering

arXiv

引用

arXiv 2025年

作者： Zhuo, Hao Yang, Yicheng Peng, Kewen Bowers College of Computing and Information Science Cornell University 410 Thurston Avenue IthacaNY14850 United States Department of Computer Science North Carolina State University 2101 Hillsborough Street RaleighNC27695 United States

Large Language Models (LLMs) have become integral to software engineering (SE), where they are increasingly used in development workflows. However, their widespread use raises concerns about the presence and propagation of toxic language—harmful or offensive content that can foster exclusionary environments. This paper provides a comprehensive review of recent research on toxicity detection and mitigation, focusing on both SE-specific and general-purpose datasets. We examine annotation and preprocessing techniques, assess detection methodologies, and evaluate mitigation strategies, particularly those leveraging LLMs. Additionally, we conduct an ablation study demonstrating the effectiveness of LLM-based rewriting for reducing toxicity. By synthesizing existing work and identifying open challenges, this review highlights key areas for future research to ensure the responsible deployment of LLMs in SE and beyond. Copyright © 2025, The Authors. All rights reserved.

关键词： Software design

来源：评论

学校读者我要写书评

暂无评论

P2P trading and Energy storage technologies for mitigating balancing challenges of RESs integration into multi-energy power systems: A new paradigm for DER integration

引用

Journal of Energy Storage 2025年 124卷

作者： Li, Yong Shaban, Mohamed Salih, Sinan Q. Almadhor, Ahmad Aoudia, Mouloud Abdullaeva, Barno Alkhalaf, Salem Albalawi, Hind Mahariq, Ibrahim Alalawi, Amr School of Business Shanghai Normal University Tianhua College Jiading Destrict Shanghai201815 China Department of Physics Faculty of Science Islamic University of Madinah Madinah42351 Saudi Arabia Technical College of Engineering Al-Bayan University Baghdad10011 Iraq Department of Computer Engineering and Networks College of Computer and Information Sciences Jouf University Saudi Arabia Department of Industrial Engineering College of Engineering Northern Border University P.O. Box 1321 Arar91431 Saudi Arabia Department of Mathematics and Information Technologies Vice-Rector for Scientific Affairs Tashkent State Pedagogical University Tashkent Uzbekistan Department of Computer Engineering College of Computer Qassim University Buraydah Saudi Arabia Department of Physics College of Sciences Princess Nourah bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia University College Korea University Seoul02481 Korea Republic of Department of Medical Research China Medical University Hospital China Medical University Taichung Taiwan Applied Science Research Center Applied Science Private University Amman Jordan Department of Mathematical Science College of Engineering University of Business and Technology Jeddah21361 Saudi Arabia

Renewable energy (RES) integration is essential for sustainability but poses grid-balancing challenges. Shifting RES uncertainty to the gas grid can facilitate balancing issues, though it raises emissions. Energy storage (e.g., power to hydrogen gas (P2G), batteries) aids balancing but requires coordinated controlling action. Peer-to-peer (P2P) markets enable local distributed energy resource (DER) integration yet must address grid constraints, such as losses, voltage stability, and emissions, to ensure cost-effective, optimal performance. In response to these concerns, this research presents a multi-objective decentralized P2P energy trading framework to facilitate local energy trading in the modern integrated energy microgrid (IEM). The proposed framework consists of two levels: the scheduling stage and the energy trading stage. At the first level, the energy scheduling of the studied IEM's customers is determined through multi-objective optimization, considering the minimization of operating costs, carbon emission, and energy loss as three objective functions. Based on the first-level results, customers, P2Gs, gas-fired generators, RESs, and prosumers initiate energy trading in the second level. The proposed multi-objective optimization problem is solved through the Ε-constraint method, and the P2P decentralized energy market is performed through a continuous double auction mechanism (CDAM). Also, the first level of the presented method is equipped with the TOPSIS framework to choose the optimum solution from the Pareto front. The proposed method is implemented on the 33-buse distribution power grid and 15-node natural gas network. The simulation results show that employing the proposed method reduces total operating cost, peak load, carbon emission, and energy loss by 20 %, 21 %, 26 %, and 58.3 % compared to without the presented optimization method. Also, the results of the Ε-constraint method are compared with the widely used NSGA-II framework. The outcomes de

关键词： Battery management systems

来源：评论

学校读者我要写书评

暂无评论

Predictive Modeling of Solar Power Generation Using Deep Learning for Real-Time Optimization in Photovoltaic System

Predictive Modeling of Solar Power Generation Using Deep Lea...

引用

2025 International Conference on Intelligent Control, Computing and Communications, IC3 2025

作者： Bafna, Minal Thakur, Hardeo Kumar Singh, Jagendra Reddy, P. Srinivas Ekvitayavetchanukul, Pongkit Rana, Manoj Agrawal P. G. College Department of Physics Jaipur India School of Computer Science Engineering & Technology Bennett University Greater Noida India Cvr College of Engineering Department of Mechanical Engineering Hyderabad India The Board of Khonkaen University Affaneirs Khonkaen University Thailand Teerthanker Mahaveer University Cdoe Moradabad India

ISBN: (纸本)9798331527495

This research addresses the need for precise solar power generation forecasting and optimization of energy management within photovoltaic systems;this is important as there is a greater demand for renewable energy sources. Reliable predictions of the output of solar power would enhance stability in the grid and thus ensure efficient distribution of the energy. This research uses deep learning techniques, the Long Short-Term memory (LSTM) model, to predict solar power generation from several environmental variables, including solar irradiation, air temperature, wind speed, relative humidity, air pressure, module power, and module temperature. To validate the performance of the model, rigorous use of metrics like Mean absolute error (MAE), Root Mean Square Error (RMSE), and Mean absolute percentage error (MAPE) was made. The proposed LSTM model gave an MAE of 0.25 kW, an RMSE of 0.35 kW, and a MAPE of 5%. This was better compared to other models like Gated recurrent unit (GRU), recurrent neural network (RNN), and Artificial neural network (ANN). A real-time implementation of the LSTM model depicted the fact that it is feasible to use the model to make correct predictions of the amount of solar power produced throughout the day. At peak periods, such as midday hours, accuracy levels hit 98% in some cases, whereas the model is compelling enough to pinpoint and respond accordingly to changing solar power production volumes for optimizing load management and energy supply. This research verifies the deep learning model and its efficiency in energy forecasting, with intense stress on the practical application of the LSTM model in real-time energy management, contributing a lot to the advancement of renewable energy integration and optimizing power systems. Finally, this research provides valuable insight into improving the operational efficiency of photovoltaic systems in order to make the future energy scenario more sustainable. © 2025 IEEE.

关键词： Power management (telecommunication)

来源：评论

学校读者我要写书评

暂无评论

An Extensive and Methodical Review of Smart Grids for Sustainable Energy Management-Addressing Challenges with AI, Renewable Energy Integration and Leading-edge Technologies

引用

IEEE Access 2025年

作者： Biswas, Parag Rashid, Abdur Al Masum, Abdullah Al Nasim, Md Abdullah Anas Ferdous, A.S.M. Gupta, Kishor Datta Biswas, Angona Westcliff University MSEM Department California United States Pioneer Alpha Research and Development Department Dhaka Bangladesh Bangladesh University of Engineering and Technology Department of Biomedical Engineering Dhaka Bangladesh Clark Atlanta University Department of Computer and Information Science GA United States

Energy management decreases energy expenditures and consumption while simultaneously increasing energy efficiency, reducing carbon emissions, and enhancing operational performance. Smart grids are a type of sophisticated energy infrastructure that increase the generation and distribution of electricity's sustainability, dependability, and efficiency by utilizing digital communication technologies. They combine a number of cutting-edge techniques and technology to improve energy resource management. A large amount of research study on the topic of smart grids for energy management has been completed in the last several years. The authors of the present study want to cover a number of topics, including smart grid benefits and components, technical developments, integrating renewable energy sources, using artificial intelligence and data analytics, cybersecurity, and privacy. Smart Grids for Energy Management are an innovative field of study aiming at tackling various difficulties and magnifying the efficiency, dependability, and sustainability of energy systems, including: 1) Renewable sources of power like solar and wind are intermittent and unpredictable 2) Defending smart grid system from various cyber-attacks 3) Incorporating an increasing number of electric vehicles into the system of power grid without overwhelming it. Additionally, it is proposed to use AI and data analytics for better performance on the grid, reliability, and energy management. It also looks into how AI and data analytics can be used to optimize grid performance, enhance reliability, and improve energy management. The authors will explore these significant challenges and ongoing research. Lastly, significant issues in this field are noted, and recommendations for further work are provided. © 2013 IEEE.

关键词： Clean energy

来源：评论

学校读者我要写书评

暂无评论

Real-Time Interaction in AR/VR Environments: A Deep Learning Approach to Human-computer Interaction

Real-Time Interaction in AR/VR Environments: A Deep Learning...

引用

Advances in computer science, Electrical, Electronics, and Communication Technologies (CE2CT), International Conference on

作者： U. Chindiyababy Vishal Mehta Preetish Kakkar Matchanova Dilbar Abdalov Umidbek Aditya Sharma Department of Information Technology SRM Valliammai Engineering College India Independent UX Professional Independent Researcher India Urgench state university Uzbekistan Department of History Mamun university Khiva Uzbekistan Department of Computer Science and Engineering Tula's Institute Dehradun India

ISBN: (数字)9798331518578

ISBN: (纸本)9798331518585

AR and VR technologies represent Innovative technologies of HCI providing rich and dynamic user experiences. This work proposes a framework based on deep learning for interacting in real-time with AR/VR environments and employing innovative multimodal input handling and flexible response strategies. The implementation consists of convolutional neural networks (CNNs) for gesture recognition, recurrent neural models for natural language understanding, and transformer-based models for gaze tracking. These models are latency-reduced and are deployed with minimal overhead architectures to fit into constrained AR/VR devices. Several assessments are done based on actual data, and mock AR/VR is used to help prototypes. Experimental evaluation confirms that the proposed framework has better interaction latency, accuracy, and adaptability over conventional approaches. For instance, gesture recognition has a success rate of 96.3% the response is facilitated in less than 50 and the natural language commands have a processing delay of 30 ms. The user surveys support the case that the system effectively presents context-based and individualized communication making the system more engaging and easier to use. It focuses on the issues that arise while implementing such methods, including computational load and integration difficulties, and provides a more detailed version of this work that describes solutions to them. The findings map the feasibility of deep learning in augmenting HCI within the AR/VR systems, which will help propel advancements across education, healthcare, and more notably, game-designing and telepresence.

关键词： Deep learning Recurrent neural networks Atmospheric modeling Education Gesture recognition Medical services Gaze tracking Real-time systems Haptic interfaces Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Optimal Bilinear Equalizer for Cell-Free Massive MIMO Systems over Correlated Rician Channels

引用

IEEE Transactions on Signal Processing 2025年

作者： Wang, Zhe Zhang, Jiayi Bjornson, Emil Niyato, Dusit Ai, Bo Beijing Jiaotong University State Key Laboratory of Advanced Rail Autonomous Operation Beijing100044 China Beijing Jiaotong University School of Electronics and Information Engineering Beijing100044 China KTH Royal Institute of Technology Department of Computer Science Stockholm114 28 Sweden Nanyang Technological University College of Computing & Data Science 639798 Singapore

In this paper, we explore the low-complexity optimal bilinear equalizer (OBE) combining scheme design for cell-free massive multiple-input multiple-output networks with spatially correlated Rician fading *** provide a spectral efficiency (SE) performance analysis framework for both the centralized and distributed processing schemes with bilinear equalizer (BE)-structure combining schemes applied. The BE-structured combining is a set of schemes that are constructed by the multiplications of channel statistics-based BE matrices and instantaneous channel estimates. Notably, we derive closed-form achievable SE expressions for centralized and distributed BE-structured combining schemes. We propose one centralized and two distributed OBE schemes: Centralized OBE (C-OBE), Distributed OBE based on Global channel statistics (DG-OBE), and Distributed OBE based on Local channel statistics (DL-OBE), which maximize their respective SE expressions. OBE matrices in these schemes are tailored based on varying levels of channel statistics. Notably, we obtain new and insightful closed-form results for the C-OBE, DG-OBE, and DL-OBE combining schemes. Numerical results demonstrate that the proposed OBE schemes can achieve excellent SE, even in scenarios with severe pilot contamination. © 2025 IEEE. All rights reserved.

关键词： Fading channels

来源：评论

学校读者我要写书评

暂无评论

Controller Optimization of an Automated Force Calibration Device Based on an Improved Particle Swarm Algorithm

Controller Optimization of an Automated Force Calibration De...

引用

Sensors and information technology (ICSI), International Conference on

作者： Jiming Jiang Xiang Dong Qiang Li Huibin Cao School of Electrical Engineering and Automation Anhui University Hefei China Department of Computer Science and Technology College of Big Data and Internet Shenzhen Technology University Shenzhen China Hefei Institutes of Physical Science Chinese Academy of Sciences Hefei Anhui China

ISBN: (数字)9798331542887

ISBN: (纸本)9798331542894

This paper presents a PID controller optimization method based on an Improved Particle Swarm Optimization (IPSO) algorithm, aimed at enhancing the design and performance of automated force calibration devices. Traditional force calibration systems often suffer from limitations in automation and force application accuracy, which this method seeks to address. The IPSO algorithm incorporates logistic chaotic mapping to initialize particle positions and enhances the standard particle swarm algorithm through nonlinear decreasing weights and adaptive learning factors, resulting in improved global search capabilities and faster local convergence. Additionally, the objective function is refined by normalizing and weighting the integral of time-multiplied absolute error (ITAE) and the overshoot, thereby further enhancing the force application accuracy of the PID controller. Simulation results demonstrate that the IPSO algorithm offers significant advantages in terms of settling time, overshoot, and ITAE error when compared to traditional PID tuning methods (e.g., Ziegler-Nichols) and other optimization algorithms (e.g., GA, GWO). Experimental results validate the effectiveness and practicality of the IPSO algorithm, providing an efficient solution for the automation and high-precision control of force sensor calibration systems.

关键词： Performance evaluation Accuracy Automation Force Optimization methods Systems modeling Calibration Particle swarm optimization Tuning Standards

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：