检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2024年第10期5卷 4959-4971页

作者： Lu, Haoquan Lai, Zhihui Zhang, Junhong Yu, Zhuozhen Wen, Jiajun Shenzhen University Computer Vision Institute College of Computer Science and Software Engineering Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen518060 China Peking University Shenzhen Graduate School School of Electronic and Computer Engineering Shenzhen518055 China

Ridge regression (RR)-based methods aim to obtain a low-dimensional subspace for feature extraction. However, the subspace's dimensionality does not exceed the number of data categories, hence compromising its capability of feature representation. Moreover, these methods with L2-norm metric and regularization cannot extract highly robust features from data with corruption. To address these problems, in this article, we propose generalized jointly sparse linear discriminant regression (GOAL), a novel regression method based on joint L2,1-norm and capped-L2-norm, which can integrate sparsity, locality, and discriminability into one model to learn a full-rank robust feature extractor. The sparsely selected discriminative features are robust enough to characterize the decision boundary between classes. Locality is related to manifold structure and Laplacian smoothing, which can enhance the robustness of the model. By using the multinorm metric and regularization regression framework, the proposed method obtains the projection with joint sparsity and guarantees that the rank of the projection matrix will not be limited by the number of classes. An iterative algorithm is proposed to compute the optimal solution. Complexity analysis and proofs of convergence are also given in the article. Experiments on well-known datasets demonstrate our model's superiority and generalization ability. © 2020 IEEE.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Multi-timescale feature extraction method of wastewater treatment process based on adaptive entropy

引用

Chinese Journal of Chemical Engineering 2024年第12期76卷 264-271页

作者： Honggui Han Yaqian Zhao Xiaolong Wu Hongyan Yang School of Information Science and Technology Beijing University of TechnologyBeijing 100124China College of Environmental Sciences and Engineering Beijing University of TechnologyBeijing 100124China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing 100124China Engineering Research Center of Digital Community Ministry of EducationBeijing 100124China

In wastewater treatment systems,extracting meaningful features from process data is essential for effective monitoring and ***,the multi-time scale data generated by different sampling frequencies pose a challenge to accurately extract *** solve this issue,a multi-timescale feature extraction method based on adaptive entropy is ***,the expert knowledge graph is constructed by analyzing the characteristics of wastewater components and water quality data,which can illustrate various water quality parameters and the network of relationships among ***,multiscale entropy analysis is used to investigate the inherent multi-timescale patterns of water quality data in depth,which enables us to minimize information loss while uniformly optimizing the ***,we harness partial least squares for feature extraction,resulting in an enhanced representation of sample data and the iterative enhancement of our expert knowledge *** experimental results show that the multi-timescale feature extraction algorithm can enhance the representation of water quality data and improve monitoring capabilities.

关键词： Feature extraction Knowledge graph Wastewater treatment process Adaptive entropy

来源：评论

学校读者我要写书评

暂无评论

Surface repair of wide-bandgap perovskites for high-performance all-perovskite tandem solar cells

引用

Journal of Energy Chemistry 2024年第6期93卷 64-70,I0003页

作者： Xiaojing Lv Weisheng Li Jin Zhang Yujie Yang Xuefei Jia Yitong Ji Qianqian Lin Wenchao Huang Tongle Bu Zhiwei Ren Canglang Yao Fuzhi Huang Yi-Bing Cheng Jinhui Tong State Key Laboratory of Advanced Technology for Materials Synthesis and Processing Wuhan University of TechnologyWuhan 430070HubeiChina Key Lab of Artificial Micro-and Nano-Structures of Ministry of Education of China School of Physics and TechnologyWuhan UniversityWuhan 430072HubeiChina Department of Electrical and Electronic Engineering Photonics Research Institute(PRI)Research Institute for Intelligent Wearable Systems(iWEAR)The Hong Kong Polytechnic University999077Hong KongChina Laboratory of Advanced Materials Fudan UniversityShanghai 200438China Xianhu Laboratory of the Advanced Energy Science and Technology Guangdong Laboratory Foshan 528200GuangdongChina

Wide-bandgap(WBG)perovskite solar cells(PSCs)play a fundamental role in perovskite-based tandem solar ***,the efficiency of WBG PSCs is limited by significant open-circuit voltage losses,which are primarily caused by surface *** this study,we present a novel method for modifying surfaces using the multifunctional S-ethylisothiourea hydrobromide(SEBr),which can passivate both Pb^(-1)and FA^(-1)terminated surfaces,Moreover,the SEBr upshifted the Fermi level at the perovskite interface,thereby promoting carrier *** proposed method was effective for both 1.67 and 1.77 eV WBG PSCs,achieving power conversion efficiencies(PCEs)of 22.47%and 19.90%,respectively,with V_(OC)values of 1.28 and 1.33 V,along with improved film and device *** this advancement,we were able to fabricate monolithic all-perovskite tandem solar cells with a champion PCE of 27.10%,This research offers valuable insights for passivating the surface trap states of WBG perovskite through rational multifunctional molecular engineering.

关键词： Wide-bandgap perovskite Surface defect Multifunctional molecule All-perovskite tandem solar cells

来源：评论

学校读者我要写书评

暂无评论

Unveiling the orientation growth mechanism and solar-blind response performance of β-Ga2O3(100)film on SiC substrate with AlN buffer layer

引用

材料科学技术（英文版） 2025年第7期210卷 20-28页

作者： Jie Su Zixin Zhang Liang Shi Liping Feng Fuchao He Jingjing Chang Jincheng Zhang Yue Hao State Key Laboratory of Wide Bandgap Semiconductor Devices and Integrated Technology School of MicroelectronicsXidian UniversityXi'an 710071PR China State Key Lab of Solidification Processing School of Materials Science and EngineeringNorthwestern Polytechnical UniversityXi'an 710072PR China State Key Laboratory of Crystal Materials Shandong UniversityJinan 250100PR China State Key Laboratory of Wide Bandgap Semiconductor Devices and Integrated Technology School of MicroelectronicsXidian UniversityXi'an 710071PR China State Key Lab of Solidification Processing School of Materials Science and EngineeringNorthwestern Polytechnical UniversityXi'an 710072PR China School of Electronics Information and Electrical Engineering Shanghai Jiao Tong UniversityShanghai 200240PR China

Optimizing the orientation of β-Ga2O3 has emerged as an effective strategy to design high-performanceβ-Ga2O3 device,but the orientation growth mechanism and approach have not been revealed ***,by employing AlN buffer layer,the highly preferred orientation of β-Ga2O3(100)film rather than(-201)film is realized on 4H-SiC substrate at low sputtering power and *** β-Ga2O3(100)film exhibits a slower growth speed than(-201)film,the former possesses the higher dangling bond density and the lower nucleation energy,and a large conversion barrier exists between these two ***,the AlN buffer layer can suppress the surface oxidation of the 4H-SiC substrate and eliminate the strain of β-Ga2O3(100)film,which further reduces the nucleation energy and en-larges the conversion ***,the AlN buffer layer can increase the oxygen vacancy formation energy and decrease the oxygen vacancy concentration of β-Ga2O3(100)***,the solar-blind photodetector based on the oriented film exhibits the outstanding detectivity of 1.22 × 1012 Jones and photo-to-dark current ratio of 1.11 × 105,which are the highest among the reported β-Ga2O3 solar-blind photodetector on the SiC *** results offer in-depth insights into the preferred orientation growth mechanism,and provide an effective way to design high-quality β-Ga2O3(100)orientation film and high-performance solar-blind photodetector.

关键词： β-Ga2O3(100)film Orientation growth AlN buffer layer Solar-blind photodetector DFT calculation

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised Speech Lie Detection Algorithm Based on Multiple Features and Adaptive Threshold

IAENG International Journal of Computer Science

引用

IAENG International Journal of Computer Science 2025年第6期52卷 2015-2021页

作者： Zheng, Shuailong Tao, Huawei Yu, Hang Xi, Ji Zhu, Chunhua Wang, Xuemei Key Laboratory for Grain Information Processing and Control Henan Key Laboratory of Grain Storage Information Intelligent Perception and Decision Making Henan University of Technology Zhengzhou450001 China Computer Information Engineering Changzhou Institute of Technology Jiangsu213000 China

The limited number of labeled samples can significantly affect the accuracy of speech lie detection systems. However, semi-supervised algorithms have been shown to effectively improve detection performance under such constraints. To address this challenge, this paper proposes a semi-supervised speech lie detection algorithm that utilizes multiple features and adaptive thresholds. First, through comparative analysis, this paper identifies 312-dimensional acoustic features and 768-dimensional self-supervised features as the key elements of lie detection. Then, these acoustic features are processed using a bidirectional long short-term memory network (Bi-LSTM), which obtains comprehensive information by analyzing acoustic statistical features in both forward and backward directions. Subsequently, this paper develops a semi-supervised feature fusion module (BSFM) to extract deeper insights from basic features and achieves adaptive feature fusion through targeted training. In addition, based on a detailed analysis of the distribution characteristics of lie detection samples, this paper proposes an improved Freematch algorithm to better utilize unlabeled data. Experimental results show that the proposed algorithm outperforms most existing methods on the self-built H-Wolf dataset and the Columbia/SRI/Colorado Corpus (CSC). © (2025), (International Association of Engineers). All Rights Reserved.

关键词： Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

AIoT for Aircraft Final Assembly:An intelligent and Collaborative Framework

引用

Journal of Communications and information Networks 2024年第3期9卷 262-276页

作者： Cheng Ren Xiaojing Wen Xinping Guan Cailian Chen Yehan Ma Department of Automation Shanghai Jiao Tong UniversityShanghai 200240China Key Laboratory of System Control and Information Processing Ministry of Education of ChinaShanghai 200240China Shanghai Engineering Research Center of Intelligent Control and Management Shanghai 200240China School of Electronic Information and Electrical Engineering Shanghai Jiao Tong UniversityShanghai 200240China

Aircraft final assembly line(AFAL)involves thousands of processes that must be completed before ***,the heavy reliance on manual labor in most assembly processes affects the quality and prolongs the delivery *** the advent of artificial intelligence of things(AIoT)technologies has introduced advancements in certain AFAL scenarios,systematically enhancing the intelligence level of the AFAL and promoting the widespread deployment of artificial intelligence(AI)technologies remain significant *** address these challenges,we propose the intelligent and collaborative aircraft assembly(ICAA)framework,which integrates AI technologies within a cloud-edge-terminal *** ICAA framework is designed to support AI-enabled applications in the AFAL,with the goal of improving assembly efficiency at both individual and multiple process *** analyze specific demands across various assembly scenarios and introduce corresponding AI technologies to meet these *** three-tier ICAA framework consists of the assembly field,edge data platform,and assembly cloud platform,facilitating the collection of heterogeneous terminal data and the deployment of AI *** framework enhances assembly efficiency by reducing reliance on manual labor for individual processes and fostering collaboration across multiple *** provide detailed descriptions of how AI functions at each level of the ***,we apply the ICAA framework to a real AFAL,focusing explicitly on the flight control system testing *** practical implementation demonstrates the effectiveness of the framework in improving assembly efficiency and promoting the adoption of AIoT technologies.

关键词： intelligent aircraft final assembly line AIoT cloud-edge-terminal framework design

来源：评论

学校读者我要写书评

暂无评论

Distributed Robust Safety-Critical Predefined-Time Control for Target Spherical Surrounding of UAVs

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on intelligent Vehicles 2024年 1-10页

作者： Huang, Yanting Han, Honggui Zhu, Ming Faculty of Information Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System Engineering Research Center of Digital Community Ministry of Education Beijing University of Technology Beijing China Institute of Unmanned System Beihang University Beijing China

This paper develops a distributed safety-critical predefined-time control method for the 3-dimensional (3D) moving target surrounding problem of multiple UAVs. A new 3D surrounding shape is first formulated for multiple UAVs to distribute on a moving spherical surface around the target. To reduce the communication and load burden, a predefined-time distributed estimator is designed for a subset of UAVs without any access to the target, so that the target's motion information can be accurately estimated within a predefined time. Subsequently, a robust safety-constrained predefined-time control law is designed with a twofold structure, where the nominal control term guarantees the predefined-time convergence under the safety constraints, and the adjunctive control term overcomes the influence of disturbances. A new dynamic parameter-based predefined-time control method is proposed, which can relax the unnecessary requirement of control design for achieving predefined-time convergence. The corresponding dynamic parameter-based predefined-time control Lyapunov function (DPP-CLF) is constructed together with second-order control barrier function (SCBF) to map the convergence requirement and safety constraints on states to constraints on control inputs. The nominal control term is obtained by minimizing the control inputs under constraints with linear programming. The adjunctive control term is acquired by combining a filter-based estimator and an integral sliding mode surface design, providing an accurate compensation for the disturbances. The theoretical analysis and simulation results demonstrate the predefined-time stability and collision avoidance performance of the multi-UAV system. IEEE

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Transfer Learning in Motor Imagery Brain Computer Interface: A Review

引用

Journal of Shanghai Jiaotong university(Science) 2024年第1期29卷 37-59页

作者：李明爱许冬芹 Faculty of Information Technology Beijing University of TechnologyBeijing 100124China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing 100124China Engineering Research Center of Digital Community Ministry of EducationBeijing 100124China

Transfer learning,as a new machine learning methodology,may solve problems in related but different domains by using existing knowledge,and it is often applied to transfer training data from another domain for model training in the case of insuficient training *** recent years,an increasing number of researchers who engage in brain-computer interface(BCI),have focused on using transfer learning to make most of the available electroencephalogram data from different subjects,effectively reducing the cost of expensive data acquisition and labeling as well as greatly improving the learning performance of the *** paper surveys the development of transfer learning and reviews the transfer learning approaches in *** addition,according to the"what to transfer"question in transfer learning,this review is organized into three contexts:instance-based transfer learning,parameter-based transfer learning,and feature-based transfer ***,the current transfer learning applications in BCI research are summarized in terms of the transfer learning methods,datasets,evaluation performance,*** the end of the paper,the questions to be solved in future research are put forward,laying the foundation for the popularization and in-depth research of transfer learning in BCI.

关键词： transfer learning brain-computer interface(BCI) electroencephalogram machine learning

来源：评论

学校读者我要写书评

暂无评论

Conditional selection with CNN augmented transformer for multimodal affective analysis

引用

CAAI Transactions on Intelligence technology 2024年第4期9卷 917-931页

作者： Jianwen Wang Shiping Wang Shunxin Xiao Renjie Lin Mianxiong Dong Wenzhong Guo College of Computer and Data Science Fuzhou UniversityFuzhouChina College of Computer and Cyber Security Fujian Normal UniversityFuzhouChina Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou UniversityFuzhouChina Digital Fujian Institute of Big Data Security Technology Fujian Normal UniversityFuzhouChina Department of Sciences and Informatics Muroran Institute of TechnologyMuroranJapan

Attention mechanism has been a successful method for multimodal affective analysis in recent years. Despite the advances, several significant challenges remain in fusing language and its nonverbal context information. One is to generate sparse attention coefficients associated with acoustic and visual modalities, which helps locate critical emotional se-mantics. The other is fusing complementary cross‐modal representation to construct optimal salient feature combinations of multiple modalities. A Conditional Transformer Fusion Network is proposed to handle these problems. Firstly, the authors equip the transformer module with CNN layers to enhance the detection of subtle signal patterns in nonverbal sequences. Secondly, sentiment words are utilised as context conditions to guide the computation of cross‐modal attention. As a result, the located nonverbal fea-tures are not only salient but also complementary to sentiment words directly. Experi-mental results show that the authors’ method achieves state‐of‐the‐art performance on several multimodal affective analysis datasets.

关键词： affective computing data fusion information fusion multimodal approaches

来源：评论

学校读者我要写书评

暂无评论

Blood pressure monitoring via double sandwich-structured triboelectric sensors and deep learning models

引用

Nano Research 2022年第6期15卷 5500-5509页

作者： Xu Ran Fangyuan Luo Zhiming Lin Zhiyuan Zhu Chuanjun Liu Bin Chen Chongqing Key Laboratory of Non-linear Circuit and Intelligent Information Processing College of Electronic and Information EngineeringSouthwest UniversityChongqing 400715China Research Laboratory U.S.E.Co.Ltd.Shibuya-kuTokyo 150-0013Japan

Real-time blood pressure monitoring is essential for the timely diagnosis and treatment of cardiovascular *** traditional prediction methods estimate blood pressure by measuring multiple sets of physiological signals with energyconsuming ***,a continuous,cuffless and self-powered blood pressure monitoring system was developed based on a new double sandwich-structured triboelectric sensor and a novel blood pressure method estimation.A pyramid-patterned sensor based on the double sandwich structure realizes a sensitivity of 0.89 V/kPa in a linear range of 0–35 kPa,which is more than twice of the conventional single electrode *** sensor processes a low pressure detection limit of 1 g andfast response time of 32 ***,it can easily capture the pulse signal at the radial ***,a novel method for estimating blood pressure using pulse waves accompanied by the user’s background information was *** method measures only one set of pulse signals and is portable.A deep learning model with multi-network structures was developed to improve the estimation *** mean absolute error and standard deviation of error for systolic and diastolic blood pressure(SBP and DBP)estimations were 3.79±5.27 and 3.86±5.18 mmHg,*** work reveals a new sensing structure of triboelectric sensors and offers a novel method for blood pressure estimation.

关键词： triboelectric nanogenerator self-powered sensor blood pressure monitoring deep learning

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：