检索结果-内蒙古大学图书馆

Multitarget Unsupervised Transfer Learning Improves Wearable-Sensor Cross-Body-Position Activity Recognition via Inter-Target Shared Representation

引用

IEEE SENSORS JOURNAL 2025年第5期25卷 9101-9112页

作者： Li, Jinxuan Kang, Peiqi Shull, Peter Bradley Shanghai Jiao Tong Univ Sch Mech Engn State Key Lab Mech Syst & Vibrat Shanghai 200240 Peoples R China

Human activity recognition (HAR) via wearable sensors is vital for enabling healthcare applications and facilitating human-computer interaction. HAR models need to adapt to arbitrary sensor placements to accommodate diverse human wearing preferences and movements. However, the traditional HAR model trained on data collected from a single body position performs poorly when applied to other positions due to significant variations in sensor signals. Furthermore, collecting sufficient labeled datasets for every possible sensor placement is typically time-consuming and expensive. We thus propose a multitarget and representation-shared domain adaptation approach (MTRS) that uses labeled data from a single source position to train a cross-domain model applicable to multiple target positions. Instead of solely focusing on the data distribution characteristics within each target position, MTRS simultaneously optimizes a shared feature subspace among target positions. This allows each target position to gain enhanced class boundary information from other target positions, significantly enhancing recognition accuracy and reducing model size. Experiments show that MTRS improved classification accuracy on the RealWorld dataset by 15.5% compared to the baseline and by 6.5% compared to existing methods without a shared representation. MTRS outperformed other methods by 3.8% to 7.5% on the PAMAP2 dataset and by 3.3% to 6.9% on the UpperBody dataset. These results indicate that MTRS can develop HAR models with flexible sensor placement in a more accurate and cost-effective manner, facilitating the transition of HAR applications from controlled laboratory environments to practical, robust real-world implementations.

关键词： Accuracy Sensors Adaptation models classification algorithms Training Human activity recognition Data models Wearable sensors Sensor placement Computational modeling Domain adaptation (DA) machine learning position-independent activity recognition single target domain wearable sensors

来源：评论

学校读者我要写书评

暂无评论

Application Research of Multisource Data Fusion and Multimodel Ensemble Methods in Aircraft Approach State Prediction

引用

IEEE SENSORS JOURNAL 2025年第5期25卷 8493-8505页

作者： Cao, Kang Zhang, Yongjie Feng, Jianfei Wang, Zhanchao Northwestern Polytech Univ Sch Civil Aviat Xian 710072 Shaanxi Peoples R China Shanghai Aircraft Design & Res Inst Shanghai 201210 Peoples R China Aviat Ind Dev Res Ctr China Beijing 100012 Peoples R China

With advances in sensing technology, situational awareness and prediction based on multisource data have become crucial for ensuring flight safety. Compared to single-model approaches, multimodel ensemble prediction algorithms offer higher accuracy and robustness, yet current ensemble strategies are insufficient for further development. Therefore, this article introduces a new multimodel ensemble prediction algorithm-the minimum consensus cost ensemble (MCCE) prediction algorithm-specifically for predicting the final approach state of civil aircraft. The MCCE algorithm incorporates the "minimum consensus cost" concept from group decision-making, quantifying the bias of individual models toward different prediction outcomes and dynamically adjusting the ensemble strategy to select the result with the highest consensus among models as the final output. This approach addresses decision bias issues in traditional ensemble models, enhancing stability and accuracy. In experiments, a fusion model was designed to maximize the silhouette coefficient, combining convolutional neural networks (CNNs) and multihead attention (MHA) mechanisms to efficiently integrate multisource data and isolate final approach state features. Through independent modeling and performance comparisons with multiple classification algorithms such as SVM and CNN, results show that the MCCE algorithm achieves approximately 99.97% accuracy, maintaining about 94.71% accuracy even under 20-dB signal-to-noise ratio (SNR) noise interference, significantly outperforming other models. Based on MCCE's consensus cost, the key parameters leading to abnormal approach states in civil aircraft can be identified, helping pilots make timely, effective flight decisions to reduce approach-phase accident risks.

关键词： Predictive models Atmospheric modeling Prediction algorithms Data models Accuracy Sensors Convolutional neural networks Computer architecture classification algorithms Accidents Aircraft approach classification group decision making multimodel ensemble multimodel fusion

来源：评论

学校读者我要写书评

暂无评论

Assessment of Electrical Wear State in AC Contactor Contacts Utilizing Unsupervised Clustering and Composite Model

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Sun, Shuguang Zhao, Yaohua Wang, Jingqin Zhang, Zhe Qiu, Wenjie Hebei Univ Technol Sch Artificial Intelligence Tianjin 300130 Peoples R China Hebei Univ Technol Prov & Ministerial Coconstruct Collaborat Innovat Tianjin 300130 Peoples R China Hebei Univ Technol State Key Lab Reliabil & Intelligence Elect Equipm Tianjin 300130 Peoples R China Hebei Univ Technol Sch Elect Engn Tianjin 300130 Peoples R China

The contact within an ac contactor plays a crucial role and is susceptible to failure. The electrical wear degradation process of contacts is characterized by dynamic changes. The accurate assessment of the wear state of contacts can improve the operational reliability of the contactor to a certain degree. To address this issue, a novel online assessment approach for determining the contact electrical wear state is introduced. This method is based on unsupervised clustering and heterogeneous ensemble learning methods. The permutation entropy (PE) of the original electrical feature sequence is initially computed to comprehensively investigate the gradual microscopic degradation characteristics. Subsequently, the key features that characterize contact wear are obtained by combining feature importance analysis and correlation analysis. Furthermore, utilizing the acquired key features, the division of contact wear state is conducted offline, while the evolution of these states is monitored online. During the offline phase, the fuzzy C-means (FCM) algorithm is utilized to autonomously divide the complete life cycle into multiple states, taking into account the fuzzy boundary, multistate, and multifactor interaction characteristics in the electrical wear degradation process of contacts. By utilizing ensemble learning, the online assessment model based on random forest, temporal convolutional network, and deep belief network (RF-TCN-DBN) is developed. The model improves the accuracy of state assessment by integrating evaluation results from numerous learners. The effectiveness and superiority of the approach in practical implementation have been evidenced through illustrative examples.

关键词： Contacts Feature extraction Degradation Mathematical models Data models Entropy classification algorithms Predictive models Ensemble learning Correlation AC contactor contacts electrical wear ensemble learning fuzzy C-means (FCM) algorithm permutation entropy (PE) state assessment

来源：评论

学校读者我要写书评

暂无评论

Acoustic NLOS Signal Recognition Based on 1-D Convolutional Neural Network With Channel Attention Mechanism

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Xu, Jian Liu, Yingying Jia, Yuanbo Guo, Guangyi Cao, Shuai Anhui Univ Inst Phys Sci & Informat Technol Hefei 230601 Peoples R China Wuhan Univ State Key Lab Informat Engn Surveying Mapping & Re Wuhan 430079 Hubei Peoples R China

The precise recognition of acoustic non-line-of-sight (NLOS) signal is essential for precise and robust position measurement in acoustic indoor localization instruments. Conventional approaches, which depend on manual feature extraction and threshold-based methods, often prove to be labor-intensive and lack flexibility. Furthermore, the inherent complexity and variability of indoor settings pose significant challenges in achieving high precision and generalizability in acoustic NLOS signal recognition. This article presents a novel 1-D convolutional neural network (1D-CNN) that integrates a channel attention mechanism (CAM) and a long short-term memory (LSTM) network to enhance acoustic NLOS signal classification. The CAM enhances prominent features by modulating the feature weights at the channel level and the LSTM network is used to capture dynamic temporal dependencies, thus ensuring precise signal recognition. The proposed model enables automated feature extraction and adapts to various indoor environments. The experimental results show that the proposed model achieves an NLOS recognition accuracy of 98.7% when using the identical dataset for both training and testing. Furthermore, the NLOS recognition accuracy attains 97.7% when evaluated with new test data acquired from various indoor environments, demonstrating superior performance compared to established benchmark classification algorithms.

关键词： Acoustics Feature extraction Accuracy Convolutional neural networks Nonlinear optics Location awareness Instruments Acoustic measurements classification algorithms Position measurement 1-D convolutional neural network (1D-CNN) acoustic indoor localization acoustic non-line-of-sight (NLOS) signal recognition channel attention mechanism (CAM) long short-term memory (LSTM)

来源：评论

学校读者我要写书评

暂无评论

A Novel Synthetic Minority Oversampling Technique for Multiclass Imbalance Problems

引用

IEEE ACCESS 2025年 13卷 6054-6066页

作者： Wang, Jiao Awang, Norhashidah Univ Sains Malaysia Sch Math Sci George Town 11800 Malaysia Puer Univ Sch Math & Stat Puer 665000 Peoples R China

Multi-class imbalanced datasets present significant challenges in many real-world classification tasks, where certain classes are severely underrepresented. This study addresses the classification problems with multi-class imbalanced datasets, which are inherently more complicated than binary imbalanced problems. To tackle this problem, a novel and effective method called the One-vs-One Center Hybrid Synthetic Minority Over-sampling Technique (OCH-SMOTE) algorithm is proposed, which combines the enhanced Synthetic Minority Oversampling Techniques (SMOTE) with the One-vs-One (OVO) decomposition strategy. The OCH-SMOTE algorithm comprises two key components: the OVO strategy is used to decompose the multi-class imbalanced datasets, and the enhanced CH-SMOTE algorithm is used to generate the balanced training datasets to improve the classification performance. The OCH-SMOTE algorithm is extensively evaluated on 18 real-world multi-class imbalanced datasets, using the CART decision tree as the base classifier. The proposed method is compared with classical and state-of-the-art oversampling methods. On average, the OCH-SMOTE algorithm improves P-macro by 8.19%, MAVA by 9.19%, MG by 30.68%, MFM by 8.78%, and Kappa coefficient by 0.0462 across all datasets compared to the baseline methods. The experimental results demonstrate that the OCH-SMOTE algorithm significantly enhances multi-class imbalanced datasets classification performance.

关键词： classification algorithms Accuracy Measurement Prediction algorithms Data mining Generative adversarial networks Predictive models Focusing Complexity theory Sampling methods Multi-class imbalanced over-sampling SMOTE algorithm one-vs-one approach

来源：评论

学校读者我要写书评

暂无评论

Complex Signal Analysis for Inter-Turn Short-Circuits Faults on Induction Motors

引用

IEEE SENSORS JOURNAL 2025年第8期25卷 13433-13440页

作者： Cardenas-Cornejo, Juan-Jose Almanza-Ojeda, Dora-Luz Gonzalez-Parada, Adrian Hernandez-Ramirez, Veronica Ibarra-Manzano, Mario-Alberto Univ Guanajuato Engn Div Irapuato Salamanca Campus Salamanca 36885 Mexico

Automatic detection of early-stage electrical faults in motors is challenging because physical signals, such as temperature or vibrations, are often linked to regular machine operation. Inter-turn short-circuit (ITSC) faults in the stator winding are a leading cause of irreversible damage. Given industrial demands, there is an ongoing need for innovative approaches that reduce dimensionality and expedite feature extraction from three-phase current signals. This work presents an ITSC multifault classification algorithm by modeling the behavior of three-phase stator currents in a complex space. To achieve this, two geometric-based and two optimization-based approaches are employed to parameterize the shape of the complex signal. Rather than directly extracting features from the raw signals, these parameters capture the similarities among incipient failures. The algorithm is evaluated using machine-learning classifiers trained on a dataset generated from an experimental test bench and a publicly available dataset. The proposed method achieved an accuracy of 95.30% across 13 categories, demonstrating its robustness and reliability and positioning it as a highly competitive alternative to state-of-the-art techniques.

关键词： Feature extraction Data models Analytical models Machine learning algorithms Stator windings Induction motors classification algorithms Circuit faults Accuracy Training Complex signal analysis fault electrical detection geometric modeling induction motors (IMs) inter-turn short-circuits (ITSCs) faults

来源：评论

学校读者我要写书评

暂无评论

Improving Identification by Removing Obstructive Minutiae From Latent Fingerprints

引用

IEEE ACCESS 2025年 13卷 7968-7985页

作者： Calderon-Calderon, Marcos Daniel Medina-Perez, Miguel Angel Monroy, Raul Tecnol Monterrey Sch Engn & Sci Ciudad Lopez Mateos 52926 Estado De Mexic Mexico Kayak Analyt Dome TowerJumeirah Lake Towers Dubai U Arab Emirates

Latent fingerprint identification is essential in forensic science for linking suspects to crime scenes or, conversely, confirming a person's innocence. However, latent fingerprints often are partial prints with undesirable characteristics such as noise or distortion. Due to these characteristics, identifying the physical details of a latent fingerprint, known as minutiae, is a complex task. Recent publications found that there are minutiae in latent fingerprints that, when removed, increase the matching score. We have defined this type of minutia as obstructive. The importance of obstructive minutiae lies in their ability to increase the identification rate when they are identified and removed. In this work, we propose a new set of features to describe obstructive minutiae in latent fingerprints. Using this set of features, we have built datasets that describe latent fingerprints from which a subset of minutiae has been removed. Additionally, we have evaluated a set of multi-class classifiers trained with our datasets to predict if there are obstructive minutiae in a latent fingerprint. Finally, we developed two novel algorithms to identify and remove, in latent fingerprints, the obstructive minutia that generates the maximum increase in the matching score according to our set of classifiers. We employed Cumulative Match Characteristic (CMC) curves to compare the relative change of identifying an initial latent fingerprint versus a latent fingerprint with the removed obstructive minutia with the maximum increase in the matching score. We achieved an identification rate of 88.2% at Rank 1 using the XGBoost classifier on the NIST SD27 database. Similarly, the Random Forest classifier achieved an identification rate of 73.5% at Rank 1 on the NIST SD302 database.

关键词： Fingerprint recognition Databases classification algorithms Prediction algorithms Bifurcation NIST Object recognition Noise Feature extraction Indexing Latent fingerprint identification minutiae classification algorithms matching score feature engineering

来源：评论

学校读者我要写书评

暂无评论

Multi-Objective Optimization for Multimodal Multi-Objective Multi-Point Shortest Path Problem Considering Unforeseeable Road Eventualities

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2025年第6期26卷 8622-8640页

作者： Xu, Zhiwei Zhang, Kai Del Ser, Javier Li, Miqing Xu, Xin He, Juanjuan Wu, Ni Wuhan Univ Sci & Technol Sch Comp Sci & Technol Hubei Prov Key Lab Intelligent Informat Proc andRe Wuhan 430065 Hubei Peoples R China Tecnalia Basque Res & Technol Alliance BRTA Bizkaia 48160 Bilbao Spain Univ Basque Country Dept Commun Engn UPV EHU Bizkaia 48013 Bilbao Spain Univ Birmingham Sch Comp Sci Birmingham B15 2TT England

Multi-objective multi-point shortest path planning problems are commonly encountered in real-world applications. Numerous path planning algorithms have been proposed to accommodate different model assumptions. However, most existing algorithms can only identify a subset of the Pareto optimal paths and overlook equivalent Pareto optimal paths. Relying solely on a subset of Pareto optimal solutions is insufficient to effectively respond to unforeseeable road eventualities in the real-world traffic environment. In this paper, multi-objective multi-point shortest path planning problem is modeled as a multimodal multi-objective optimization problem with necessary points constrains. A multimodal multi-objective evolutionary algorithm using constraint dominance principle-based path comparison strategy and path similarity-based multimodal solutions selection strategy is proposed to address this problem. The proposed constraint dominance principle-based path comparison strategy can effectively navigate through large infeasible regions by relaxing necessary point constraints, thereby obtaining a true constrained Pareto front. The proposed path similarity-based multimodal solutions selection strategy can effectively balance the distribution of solutions in the decision space, thereby preserving multiple equivalent optimal solutions. The proposed algorithm is compared with five state-of-the-art path planning algorithms from the benchmark test suite derived from the 2021 IEEE CEC path planning competition, where city maps are adapted from real transportation networks in Chinese cities, in our experiments. The exceptional performance is demonstrated through thirty independent runs, yielding experimental results that showcase the superiority of the proposed algorithm on the test problem set. This superior performance highlights the potential for designing more resilient path planners suitable for scenarios affected by unpredictable road eventualities.

关键词： Path planning Optimization classification algorithms Shortest path problem Pareto optimization Search problems Heuristic algorithms Roads Evolutionary computation Convergence Multi-objective shortest path planning constrained multi-objective optimization multimodal multi-objective optimization multi-objective evolutionary algorithm

来源：评论

学校读者我要写书评

暂无评论

Dynamic Task Assignment of Multiple Heterogeneous Autonomous Aerial Vehicles Based on Consensus With Uncertainties

引用

IEEE AEROSPACE AND ELECTRONIC SYSTEMS MAGAZINE 2025年第3期40卷 48-60页

作者： Wu, Weinan Lu, Zeyu Sun, Yiming Ang, Marcelo H. Gong, Chunlin Northwestern Polytech Univ Sch Astronaut Xian 710072 Peoples R China Natl Univ Singapore Coll Design & Engn Singapore 117608 Singapore

This study addresses the problem of maximizing the task-assignment reward of a fleet of heterogeneous autonomous aerial vehicles (AAVs) in a dynamic reconnaissance and confirmation task in uncertain scenarios and multi-AAV tasks, where the coupled path optimization objectives need to be considered. The existing consensus-based bundle algorithm is extended using an effective method for managing multitask and multiagent constraints. In addition, the Bayesian estimation is adopted to handle uncertainties in a given scenario. The proposed method is verified by the sample run tests on a disaster area reconnaissance and confirmation task. The test results verify both the practicality and advantages of the proposed method. Finally, a robust extension to the consensus-based bundle algorithm that handles coupling with the path planning optimization in dynamic search and rescue scenarios, including tasks with multi-AAV service requirements and time-critical constraints, is introduced.

关键词： Resource management Uncertainty Autonomous aerial vehicles Vehicle dynamics Time factors Reconnaissance Aerodynamics Heuristic algorithms Risk management classification algorithms Task analysis

来源：评论

学校读者我要写书评

暂无评论

Optimization of Depth Map Intra Coding algorithms Based on Machine Learning

引用

IEEE ACCESS 2025年 13卷 5846-5857页

作者： Song, Wenjun Yang, Jiaying Zhang, Qiuwen Zhengzhou Univ Light Ind Coll Comp & Commun Engn Zhengzhou 450002 Peoples R China

As a cutting-edge field of global research, 3D video technology faces the dual challenges of large data volumes and high processing complexity. Although the most recent video coding standard VVC, surpasses HEVC in coding efficiency, dedicated research on 3D video coding remains relatively scarce. Building on existing research, this study aims to develop a 3D video coding algorithm based on VVC that lowers the complexity of the encoding procedure. We focus specifically on the depth maps in 3D video content and introduce an extreme forest model from machine learning to optimize intra-frame coding. This paper proposes a novel CU partitioning strategy implemented through a two-stage extreme forest model. First, the initial model predicts the CU partitioning type, including no partition, QuadTree partitioning, Multi-type tree horizontal partitioning, and Multi-type tree vertical partitioning. For the latter two cases, a second model further refines the partitioning into binary or ternary trees. Through this two-stage prediction mechanism, we effectively bypass CU partitioning types with low probability, significantly reducing the coding complexity. The experimental results demonstrate that the proposed algorithm saves 47.46% in encoding time while maintaining coding quality, with only a 0.26% increase in Bjontegaard Delta Bitrate. This achievement provides an effective low-complexity solution for the 3D video coding field.

关键词： Encoding Prediction algorithms Complexity theory Video coding Partitioning algorithms Three-dimensional displays Forestry Predictive models Adaptation models classification algorithms Depth map intra-frame coding extreme forest

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：