检索结果-内蒙古大学图书馆

Graph Laplacian-Improved Convolutional Residual autoencoder for Unsupervised Human Action and Emotion Recognition

IEEE ACCESS 2022年 10卷 131128-131143页

作者： Paoletti, Giancarlo Beyan, Cigdem Del Bue, Alessio Fdn Ist Italiano Tecnol IIT Pattern Anal & Comp Vis PAVIS Res Line Genoa Italy Fdn Ist Italiano Tecnol IIT Pattern Anal & Comp Vis PAVIS Res Line Genoa Italy Fdn Ist Italiano Tecnol IIT Pattern Anal & Comp Vis PAVIS Res Line Genoa Italy

This paper presents an unsupervised feature learning approach based on 3D-skeleton data for human action and human discrete emotion recognition. Relying on the time series of skeleton data analysis to perform such tasks is effective and important to preserve the individual's privacy better. Besides, such methods can represent a viable alternative to emotion recognition applications, in which most works use frontal or profile facial images disclosing the subject's appearance. On the other hand, current unsupervised methods are able to encode the high variety of contexts and nature of the data, but often at the expense of a higher model complexity or longer computational time. To lessen these shortcomings, this paper proposes a convolutional residual autoencoder that models the skeletal geometry across the temporal dynamics of the data without relying on computationally expensive recurrent architectures. Our approach also implements a Graph Laplacian Regularization leveraging upon the implicit skeleton joints connectivity, further improving the robustness of the feature embeddings learned without using action or emotion labels. It was validated on large-scale datasets, having variability in the domain, the input skeleton data (e.g. the number of joints, adjacency matrices), and sensor technology. The results show its effectiveness by notably surpassing the performance of the state-of-the-art unsupervised methods while also achieving better recognition scores compared to the several fully supervised approaches. Extensive experimental analysis proves the usefulness of the proposed method under various evaluation protocols with observed higher-quality feature representations, even if when it is trained with fewer data. The results highlight the proposed method's remarkable transfer-ability across various domains, and its faster inference time.

关键词： Action recognition autoencoder emotion recognition full-body movement graph Laplacian skeletal data unsupervised feature learning

来源：评论

学校读者我要写书评

暂无评论

A Graph Attention-Based autoencoder for Critical Path Anomaly Detection in Microservices 15

A Graph Attention-Based Autoencoder for Critical Path Anomal...

引用

15th International Conference on Information and Knowledge Technology, IKT 2024

作者： Naderi, Mahdi Momeni, Hossein Shahini, Shayan Computer Engineering Department Golestan University Gorgan Iran

ISBN: (纸本)9798331522254

In complex microservice architectures, detecting performance anomalies is a critical challenge for ensuring system stability and efficiency. This study introduces CPAnoGAT (Critical Path Anomaly detection with Graph Attention Network and autoencoder), a novel real-time anomaly detection model leveraging causal graphs and critical path analysis. The model utilizes Graph Attention Networks (GAT) and a critical-path-based edge weighting strategy to focus on crucial system relationships. Features such as operation name, duration, and status codes are embedded as graph nodes, with interrelations modeled for optimal analysis. Experiments on the TrainTicket dataset, a benchmark for microservice architectures, demonstrate that CPAnoGAT outperforms state-of-the-art models such as TraceAnomaly and MultimodalTrace, achieving superior metrics including 99.98% precision, 73.38% recall, and an F1 score of 0.8464. By reducing false positives and enhancing accuracy, CPAnoGAT provides a robust tool for monitoring distributed systems. Future directions include evaluating the model on diverse datasets and integrating real-time anomaly detection in streaming environments. © 2024 IEEE.

关键词： Anomaly Detection autoencoder Critical Path Analysis Distributed Tracing Graph Attention Network Microservice

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Robust Adaptive Inverse Control of Nonlinear Dynamic Systems: Improved Settling Time with an autoencoder

引用

SENSORS 2022年第16期22卷 1401-1425页

作者： Alwan, Nuha A. S. Hussain, Zahir M. Univ Baghdad Coll Engn Baghdad 10017 Iraq Edith Cowan Univ Sch Engn Joondalup WA 6027 Australia

An adaptive deep neural network is used in an inverse system identification setting to approximate the inverse of a nonlinear plant with the aim of constituting the plant controller by copying to the latter the weights and architecture of the converging deep neural network. This deep learning (DL) approach to the adaptive inverse control (AIC) problem is shown to outperform the adaptive filtering techniques and algorithms normally used in adaptive control, especially when in nonlinear plants. The deeper the controller, the better the inverse function approximation, provided that the nonlinear plant has an inverse and that this inverse can be approximated. Simulation results prove the feasibility of this DL-based adaptive inverse control scheme. The DL-based AIC system is robust to nonlinear plant parameter changes in that the plant output reassumes the value of the reference signal considerably faster than with the adaptive filter counterpart of the deep neural network. The settling and rise times of the step response are shown to improve in the DL-based AIC system.

关键词： deep learning adaptive inverse control robust control nonlinear plant autoencoder sensor control

来源：评论

学校读者我要写书评

暂无评论

Multi-subband fusion algorithm based on autoencoder

引用

IET SIGNAL PROCESSING 2022年第9期16卷 1071-1084页

作者： Jiang, Yilin Zhang, Liting Li, Li Li, Jinxin Yang, Yaozu Harbin Engn Univ Coll Informat & Commun Engn Harbin 150001 Peoples R China Dalian Naval Acad Dept Informat Syst Dalian Peoples R China

A novel algorithm for multi-subband signal fusion achieves performance superior to traditional all-pole model, matrix pencil algorithm and deep-neural-network (Deep neural network (DNN)). The method uses a deep-learning autoencoder more fully described as a multi-subband fusion autoencoder (MSFAE). This autoencoder comprises two parts: a multi-subband encoder and a full-band decoder. Full-band echo distance envelopes are used as training data for the full-band autoencoder, to obtain the full-band coding and the full-band decoder. Then, the multi-subband echo distance envelopes are used as training data, and the full-band coding is used as labels, to train the multi-subband encoder. Finally, the multi-subband encoder and the full-band decoder are combined to obtain the MSFAE. The multi-subband distance envelopes are input to the MSFAE to obtain the full-band distance envelopes, improving the radar distance resolution and obtaining high-resolution range profiles. In contrast with the traditional all-pole model and matrix pencil algorithm, the authors' MSFAE directly processes the information in the frequency domain, avoiding the error of pole estimation in the echo domain. In contrast with DNN, the authors' MSFAE needs only multi-subband distance envelopes as input, avoiding noise subband redundancy. The experimental results show that the fusion accuracy of MSFAE is higher than the traditional all-pole model, matrix pencil algorithm and DNN. The MSFAE has superior performance using the fusion method, even at low signal-to-noise ratio.

关键词： autoencoder deep learning HRRP neural network subband fusion

来源：评论

学校读者我要写书评

暂无评论

SILP-autoencoder for face de-occlusion

引用

NEUROCOMPUTING 2022年第0期485卷 47-56页

作者： Sun, Dengdi Xie, Wandong Ding, Zhuanlian Tang, Jin Anhui Univ Sch Artificial Intelligence Key Lab Intelligent Comp & Signal Proc ICSP Minist Educ Hefei 230601 Peoples R China Anhui Univ Sch Comp Sci & Technol Anhui Prov Key Lab Multimodal Cognit Comp Hefei 230601 Peoples R China Anhui Univ Sch Internet Hefei 230039 Peoples R China

Recognizing faces with partial occlusion is a challenging problem in many real-world applications. Although various methods have been proposed to deal with the facial image de-occlusion tasks, most of them only concern the local features of occluded images, obviously ignoring the global facial expressions and structural prior information. In this paper, we propose a novel end-to-end SILP-autoencoder to effectively restore partial occluded faces. To improve the recovery quality and occlusion removal robustness, our framework mainly consists of two components, Laplacian prior subnetwork, and left-and-right symmetric match module (LR-match module), which preserve the global facial expression features and fully make use of the symmetrical characteristics of facial regions and structures respectively. Based on the above characteristics, a composite loss function is designed to achieve end-to-end training of the entire network. Extensive experiments on the face expression datasets with various shaded areas suggest that our approach achieves superior performance against the state-of-the-art methods. In particular, our method is more useful for facial detail recovery and distortion expression suppression. (c) 2022 Elsevier B.V. All rights reserved.

关键词： Face de-occlusion autoencoder Facial expression

来源：评论

学校读者我要写书评

暂无评论

Sparse autoencoder-based feature extraction from TOF-SIMS image data of human skin structures

引用

ANALYTICAL AND BIOANALYTICAL CHEMISTRY 2022年第2期414卷 1177-1186页

作者： Matsuda, Kazuhiro Aoyagi, Satoka Toray Res Ctr Ltd Surface Sci Labs 3-3-7 Sonoyama Otsu Shiga 5208567 Japan Seikei Univ Fac Sci & Technol 3-3-1 Kichijoji Kitamachi Musashino Tokyo 1808633 Japan

Time-of-flight secondary ion mass spectrometry (TOF-SIMS) is a useful and versatile tool for surface analysis, enabling detailed compositional information to be obtained for the surfaces of diverse samples. Furthermore, in the case of two- or three-dimensional imaging, the measurement sensitivity in the higher molecular weight range can be improved by using a cluster ion source, thus further enriching the TOF-SIMS information. Therefore, appropriate analytical methods are required to interpret this TOF-SIMS data. This study explored the capabilities of a sparse autoencoder, a feature extraction method based on artificial neural networks, to process TOF-SIMS image data. The sparse autoencoder was applied to TOF-SIMS images of human skin keratinocytes to extract the distribution of endogenous intercellular lipids and externally penetrated drugs. The results were compared with those obtained using principal component analysis (PCA) and multivariate curve resolution (MCR), which are conventionally used for extracting features from TOF-SIMS data. This confirmed that the sparse autoencoder matches, and often betters, the feature extraction performance of conventional methods, while also offering greater flexibility.

关键词： TOF-SIMS autoencoder Feature extraction Artificial neural network Bioimaging

来源：评论

学校读者我要写书评

暂无评论

Residual Dense autoencoder Network for Nonlinear Hyperspectral Unmixing

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2022年 15卷 5580-5595页

作者： Yang, Xu Chen, Jianguo Wang, Chengbin Chen, Zihao China Univ Geosci State Key Lab Geol Proc & Mineral Resources Wuhan 430074 Peoples R China China Univ Geosci Sch Earth Resources Wuhan 430074 Peoples R China Wuhan Univ Sch Remote Sensing & Informat Engn Wuhan 430074 Peoples R China

Hyperspectral unmixing is a popular research topic in hyperspectral processing, aiming at obtaining the ground features contained in the mixed pixels and their proportion. Recently, nonlinear mixing models have received particular attention in hyperspectral decomposition since the linear mixing model cannot suitably apply in the situation that exists in multiple scattering. In this article, we constructed a residual dense autoencoder network (RDAE) for nonlinear hyperspectral unmixing in multiple scattering scenarios. First, an encoder was built based on the residual dense network (RDN) and attention layer. The RDN is employed to characterize multiscale representations, which are further transformed with the attention layer to estimate the abundance maps. Second, we designed a decoder based on the unfolding of a generalized bilinear model to extract endmembers and estimate their second-order scattering interactions. Comparative experiments between the RDAE and six other state-of-the-art methods under synthetic and real hyperspectral datasets demonstrate that the proposed method achieved a better performance in terms of endmember extraction and abundance estimation.

关键词： Hyperspectral imaging Feature extraction Decoding Spatial resolution Scattering Licenses Kernel autoencoder generalized bilinear model (GBM) nonlinear hyperspectral unmixing residual dense network (RDN)

来源：评论

学校读者我要写书评

暂无评论

Towards latent space based manipulation of elastic rods using autoencoder models and robust centerline extractions

引用

ADVANCED ROBOTICS 2022年第3期36卷 101-115页

作者： Qi, Jiaming Ma, Guangfu Zhou, Peng Zhang, Haibo Lyu, Yueyong Navarro-Alarcon, David Harbin Inst Technol Sch Astronaut Harbin Peoples R China Hong Kong Polytech Univ Dept Mech Engn Hung Hom Hong Kong Peoples R China Beijing Inst Control Engn Natl Def Key Lab Space Intelligent Control Techno Beijing Peoples R China

The automatic shape control of deformable objects is a challenging (and currently hot) manipulation problem due to their high-dimensional geometric features and complex physical properties. In this study, a new methodology to manipulate elastic rods automatically into 2D desired shapes is presented. An efficient vision-based controller that uses a deep autoencoder network is designed to compute a compact representation of the object's infinite-dimensional shape. An online algorithm that approximates the sensorimotor mapping between the robots configuration and the object's shape features is used to deal with the latters (typically unknown) mechanical properties. The proposed approach computes the rods centerline from raw visual data in real-time by introducing an adaptive algorithm on the basis of a self-organizing network. Its effectiveness is thoroughly validated with simulations and experiments.

关键词： Robotics visual servoing deformable objects autoencoder self-organizing network

来源：评论

学校读者我要写书评

暂无评论

autoencoder-based Photoplethysmography (PPG) signal reliability enhancement in construction health monitoring

引用

AUTOMATION IN CONSTRUCTION 2024年 165卷

作者： Gautam, Yogesh Jebelli, Houtan Univ Illinois Urbana Campaign Dept Civil & Environm Engn Champaign IL 61820 USA

Prior research has validated Photoplethysmography (PPG) as a promising biomarker for assessing stress factors in construction workers, including physical fatigue, mental stress, and heat stress. However, the reliability of PPG as a stress biomarker in construction workers is hindered by motion artifacts (MA) - distortions in blood volume pulse measurements caused by sensor movement. This paper develops a deep convolutional autoencoder-based framework, trained to detect and reduce MA in MA-contaminated PPG signals. The framework's performance is evaluated using PPG signals acquired from individuals engaged in specific construction tasks. The results demonstrate the framework has effectiveness in both detecting and reducing MA in PPG signals with a detection accuracy of 93% and improvement in signal-to-noise ratio by over 88%. This research contributes to a more reliable and error-reduced usage of PPG signals for health monitoring in the construction industry.

关键词： Construction worker health monitoring Convolutional neural network Photoplethysmography Motion artifact autoencoder Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Research on Weather Process Similarity Based on 3D CNN autoencoder 3

Research on Weather Process Similarity Based on 3D CNN Autoe...

引用

3rd IEEE International Conference on Electrical Engineering, Big Data and Algorithms, EEBDA 2024

作者： Cheng, Xingguo Zhou, Yajing Li, Zeping School of Software Engineering South China University of Technology Guangzhou China

ISBN: (纸本)9798350380989

This article proposes a weather process similarity and retrieval algorithm based on a 3D CNN autoencoder. It uses 3D CNN to extract features of weather elements in multi-dimensional space. The encoder outputs a one-dimensional vector of size 512. The cosine similarity between two weather processes is then calculated based on the two one-dimensional vectors, representing their similarity. Experimental data demonstrates the effectiveness of this algorithm. By searching historical weather processes based on the forecast date and referring to the evolution of historical weather processes, it provides positive guidance for current weather forecasts. © 2024 IEEE.

关键词： autoencoder cosine similarity decoder encoder weather process similarity

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：