检索结果-南通市图书馆

A Novel Tensor Decomposition-Based Efficient Detector for Low-Altitude Aerial Objects With Knowledge Distillation Scheme

引用

IEEE/CAA Journal of Automatica Sinica 2024年第2期11卷 487-501页

作者： Nianyin zeng xinyu li Peishu Wu Han li xin Luo Department of Instrumental and Electrical Engineering Xiamen UniversityXiamen 361005China IEEE College of computer and Information science Southwest UniversityChongqing 400715China

Unmanned aerial vehicles(UAVs) have gained significant attention in practical applications, especially the low-altitude aerial(LAA) object detection imposes stringent requirements on recognition accuracy and computational resources. In this paper, the LAA images-oriented tensor decomposition and knowledge distillation-based network(TDKD-Net) is proposed,where the TT-format TD(tensor decomposition) and equalweighted response-based KD(knowledge distillation) methods are designed to minimize redundant parameters while ensuring comparable performance. Moreover, some robust network structures are developed, including the small object detection head and the dual-domain attention mechanism, which enable the model to leverage the learned knowledge from small-scale targets and selectively focus on salient features. Considering the imbalance of bounding box regression samples and the inaccuracy of regression geometric factors, the focal and efficient IoU(intersection of union) loss with optimal transport assignment(F-EIoU-OTA)mechanism is proposed to improve the detection accuracy. The proposed TDKD-Net is comprehensively evaluated through extensive experiments, and the results have demonstrated the effectiveness and superiority of the developed methods in comparison to other advanced detection algorithms, which also present high generalization and strong robustness. As a resource-efficient precise network, the complex detection of small and occluded LAA objects is also well addressed by TDKD-Net, which provides useful insights on handling imbalanced issues and realizing domain adaptation.

关键词： Attention mechanism knowledge distillation(KD) object detection tensor decomposition(TD) unmanned aerial vehicles(UAVs)

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Solving A Kind of High Complexity Multi-Objective Problems by A Fast Algorithm

引用

wuhan university Journal of Natural sciences 2003年第S1期8卷 183-188页

作者： zeng san-you, ding li-xin, kang li-shandepartment of computer science,china university of geosciences, wuhan 43.074, hubei, china Department of computer science, Zhuzhou Institute of Technology , Zhuzhou 412008, Hunan, china State Key Laboratory of Software Engineering, wuhan university, wuhan 430072, hubei, china china university of geosciences Department of Computer Science Wuhan Hubei China (GRID:grid.162107.3) (ISNI:***) Zhuzhou Institute of Technology Department of Computer Science Zhuzhou Hunan China (GRID:grid.162107.3) china university of geosciences Department of Computer Science Wuhan Hubei China (GRID:grid.162107.3) (ISNI:***) wuhan university State Key Laboratory of Software Engineering Wuhan Hubei China (GRID:grid.49470.3e) (ISNI:***)

A fast algorithm is proposed to solve a kind of high complexity multi-objective problems in this paper. It takes advantages of both the orthogonal design method to search evenly, and the statistical optimal method to speed up the computation. It is very suitable for solving high complexity problems, and quickly yields solutions which converge to the Pareto-optimal set with high precision and uniform distribution. Some complicated multi-objective problems are solved by the algorithm and the results show that the algorithm is not only fast but also superior to other MOGAS and MOEAs, such as the currently efficient algorithm SPEA, in terms of the precision, quantity and distribution of solutions.

关键词： evolutionary algorithms orthogonal design multi-objective optimization Pareto-optimal set

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Integrated ribosome and proteome analyses reveal insights into sevoflurane-induced long-term social behavior and cognitive dysfunctions through ADNP inhibition in neonatal mice

引用

Zoological Research 2024年第3期45卷 663-678页

作者： li-Rong liang Bing liu Shu-Hui Cao you-Yi Zhao Tian zeng Mei-Ting Zhai Ze Fan Dan-Yi He san-xin Ma Xiao-Tong Shi Yao Zhang Hui Zhang State Key Laboratory of Oral&Maxillofacial Reconstruction and Regeneration National Clinical Research Center for Oral DiseasesShaanxi Engineering Research Center for Dental Materials and Advanced ManufactureDepartment of AnesthesiologySchool of StomatologyAir Force Medical UniversityXi’anShaanxi 710032China Department of Anesthesiology 986th Air Force HospitalAir Force Medical UniversityXi’anShaanxi 710032China

A growing number of studies have demonstrated that repeated exposure to sevoflurane during development results in persistent social abnormalities and cognitive ***,an active fragment of the activity-dependent neuroprotective protein(ADNP),has been implicated in social and cognitive ***,the potential of davunetide to attenuate social deficits following sevoflurane exposure and the underlying developmental mechanisms remain poorly *** this study,ribosome and proteome profiles were analyzed to investigate the molecular basis of sevoflurane-induced social deficits in neonatal *** neuropathological basis was also explored using Golgi staining,morphological analysis,western blotting,electrophysiological analysis,and behavioral *** indicated that ADNP was significantly down-regulated following developmental exposure to *** adulthood,anterior cingulate cortex(ACC)neurons exposed to sevoflurane exhibited a decrease in dendrite number,total dendrite length,and spine ***,the expression levels of Homer,PSD95,synaptophysin,and vglut2 were significantly reduced in the sevoflurane ***-clamp recordings indicated reductions in both the frequency and amplitude of miniature excitatory postsynaptic currents(mEPSCs).Notably,davunetide significantly ameliorated the synaptic defects,social behavior deficits,and cognitive impairments induced by *** analysis revealed that loss of ADNP led to dysregulation of Ca^(2+)activity via the Wnt/β-catenin signaling,resulting in decreased expression of synaptic *** of Wnt signaling was restored in the davunetide-treated ***,ADNP was identified as a promising therapeutic target for the prevention and treatment of neurodevelopmental toxicity caused by general *** study provides important insights into the mechanisms underlying social and cognitive disturbances caused by sevoflurane exposure in neonatal

关键词： Davunetide Sevoflurane Abnormal social behaviors ADNP Neurotoxicity

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

MusicFace: Music-driven expressive singing face synthesis

引用

Computational Visual Media 2024年第1期10卷 119-136页

作者： Pengfei liu Wenjin Deng Hengda li Jintai Wang Yinglin Zheng Yiwei ding Xiaohu Guo Ming zeng School of Informatics Xiamen UniversityXiamen361000China Department of computer science The University of Texas at DallasRichardsonTexas75080-3021USA

It remains an interesting and challenging problem to synthesize a vivid and realistic singing face driven by music. In this paper, we present a method for this task with natural motions for the lips, facial expression, head pose, and eyes. Due to the coupling of mixed information for the human voice and backing music in common music audio signals, we design a decouple-and-fuse strategy to tackle the challenge. We first decompose the input music audio into a human voice stream and a backing music stream. Due to the implicit and complicated correlation between the two-stream input signals and the dynamics of the facial expressions, head motions, and eye states, we model their relationship with an attention scheme, where the effects of the two streams are fused seamlessly. Furthermore, to improve the expressivenes of the generated results, we decompose head movement generation in terms of speed and direction, and decompose eye state generation into short-term blinking and long-term eye closing, modeling them separately. We have also built a novel dataset, SingingFace, to support training and evaluation of models for this task, including future work on this topic. Extensive experiments and a user study show that our proposed method is capable of synthesizing vivid singing faces, qualitatively and quantitatively better than the prior state-of-the-art.

关键词： face synthesis singing music generative adversarial network

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

In-situ fabrication of on-chip 1T'-MoTe_(2)/Ge Schottky junction photodetector for self-powered broadband infrared imaging and position sensing

引用

Nano Research 2024年第6期17卷 5587-5594页

作者： Menglei Zhu Kunxuan liu Di Wu Yunrui Jiang Xue li Pei lin Zhifeng Shi xinjian li Ran ding Yalun Tang Xuechao Yu Longhui zeng School of Physics and Microelectronics and Key Laboratory of Material PhysicsZhengzhou UniversityZhengzhou 450052China Department of Electrical and computer Engineering University of California San DiegoLa JollaCA 92093USA State Key Laboratory of Integrated Optoelectronics College of Electronic Science and EngineeringJilin UniversityChangchun 130012China Key Laboratory of Multifunctional Nanomaterials and Smart Systems Suzhou Institute of Nano-Tech and Nano-BionicsChinese Academy of SciencesSuzhou 215123China

High-sensitivity room-temperature multi-dimensional infrared(IR)detection is crucial for military and civilian ***,the gapless electronic structures and unique optoelectrical properties have made the two-dimensional(2D)topological semimetals promising candidates for the realization of multifunctional optoelectronic ***,we demonstrated the in-situ construction of high-performance 1T’-MoTe_(2)/Ge Schottky junction device by inserting an ultrathin AlOx passivation *** good detection performance with an ultra-broadband detection wavelength range of up to 10.6 micron,an ultrafast response time of~160 ns,and a large specific detectivity of over 109 Jones in mid-infrared(MIR)range surpasses that of most 2D materials-based IR sensors,approaching the performance of commercial IR *** on-chip integrated device arrays with 64 functional detectors feature high-resolution imaging capability at room *** these outstanding detection features have enabled the demonstration of position-sensitive detection *** demonstrates an exceptional position sensitivity of 14.9 mV/mm,an outstanding nonlinearity of 6.44%,and commendable trajectory tracking and optoelectronic demodulation *** study not only offers a promising route towards room-temperature MIR optoelectronic applications,but also demonstrates a great potential for application in optical sensing systems.

关键词： MoTe_(2) broadband photodetection Schottky junction imaging position sensitive detector

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

EmotionIC: emotional inertia and contagion-driven dependency modeling for emotion recognition in conversation

引用

science china(Information sciences) 2024年第08期67卷 130-146页

作者： Yingjian liU Jiang li Xiaoping WANG Zhigang zeng School of Artificial Intelligence and Automation Huazhong University of Science and Technology hubei Key Laboratory of Brain-inspired Intelligent Systems Huazhong University of Science and Technology Key Laboratory of Image Processing and Intelligent Control (Huazhong university of science and Technology) Ministry of Education Institute of Artificial Intelligence Huazhong University of Science and Technology

Emotion recognition in conversation(ERC) has attracted growing attention in recent years as a result of the advancement and implementation of human-computer interface technologies. In this paper,we propose an emotional inertia and contagion-driven dependency modeling approach(EmotionIC) for ERC tasks. Our EmotionIC consists of three main components, i.e., identity masked multi-head attention(IMMHA), dialogue-based gated recurrent unit(DiaGRU), and skip-chain conditional random field(SkipCRF).Compared to previous ERC models, EmotionIC can model a conversation more thoroughly at both the feature-extraction and classification levels. The proposed model attempts to integrate the advantages of attention-and recurrence-based methods at the feature-extraction level. Specifically, IMMHA is applied to capture identity-based global contextual dependencies, while Dia GRU is utilized to extract speaker-and temporal-aware local contextual information. At the classification level, SkipCRF can explicitly mine complex emotional flows from higher-order neighboring utterances in the conversation. Experimental results show that our method can significantly outperform the state-of-the-art models on four benchmark *** ablation studies confirm that our modules can effectively model emotional inertia and contagion.

关键词： emotion recognition in conversation emotional inertia and contagion multi-head attention gated recurrent unit conditional random field

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Metal–Organic Gel Leading to Customized Magnetic‑Coupling Engineering in Carbon Aerogels for Excellent Radar Stealth and Thermal Insulation Performances

引用

Nano-Micro Letters 2024年第3期16卷 36-52页

作者： xin li Ruizhe Hu Zhiqiang Xiong Dan Wang Zhixia Zhang Chongbo liu Xiaojun zeng Dezhi Chen Renchao Che Xuliang Nie Key Laboratory of Jiangxi Province for Persistent Pollutants Control and Resources Recycle Nanchang Hangkong UniversityNanchang 330063People’s Republic of China School of Environmental and Chemical Engineering Nanchang Hangkong UniversityNanchang 330063People’s Republic of China Laboratory of Advanced Materials Department of Materials Science and Collaborative Innovation Center of Chemistry for Energy Materials and Collaborative Innovation Center of Chemistry for Energy MaterialsFudan UniversityShanghai 200438People’s Republic of China School of Materials science and Engineering Jingdezhen Ceramic UniversityJingdezhen 333403People’s Republic of China College of Chemistry and Materials Jiangxi Agricultural UniversityNanchang 330045People’s Republic of China

Metal–organic gel(MOG)derived composites are promising multi-functional materials due to their alterable composition,identifiable chemical homogeneity,tunable shape,and porous ***,stable metal–organic hydrogels are prepared by regulating the complexation effect,solution polarity and curing ***,collagen peptide is used to facilitate the fabrication of a porous aerogel with excellent physical properties as well as the homogeneous dispersion of magnetic particles during ***,two kinds of heterometallic magnetic coupling systems are obtained through the application of Kirkendall ***/nitrogen-doped carbon(NC)aerogel demonstrates an ultra-strong microwave absorption of−85 dB at an ultra-low loading of 5%.After reducing the time taken by atom shifting,a FeCo/Fe3O4/NC aerogel containing virus-shaped particles is obtained,which achieves an ultra-broad absorption of 7.44 GHz at an ultra-thin thickness of 1.59 mm due to the coupling effect offered by dual-soft-magnetic ***,both aerogels show excellent thermal insulation property,and their outstanding radar stealth performances in J-20 aircraft are confirmed by computer simulation *** formation mechanism of MOG is also discussed along with the thermal insulation and electromagnetic wave absorption mechanism of the aerogels,which will enable the development and application of novel and lightweight stealth coatings.

关键词： Metal-organic gels Heterometallic magnetic coupling Radar stealth Thermal insulation computer simulation technology

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Learning Sequential and Structural Dependencies Between Nucleotides for RNA N6-Methyladenosine Site Identification

引用

IEEE/CAA Journal of Automatica Sinica 2024年第10期11卷 2123-2134页

作者： Guodong li Bowei Zhao Xiaorui Su Dongxu li Yue Yang Zhi zeng Lun Hu xinjiang Technical Institute of Physics and Chemistry Chinese Academy of ScienceUrumqi 830011China College of computer Xi’an Jiaotong UniversityXi’an 710049China IEEE

N6-methyladenosine(m6A)is an important RNA methylation modification involved in regulating diverse biological processes across multiple ***,the identification of m6A modification sites provides valuable insight into the biological mechanisms of complex diseases at the post-transcriptional *** a variety of identification algorithms have been proposed recently,most of them capture the features of m6A modification sites by focusing on the sequential dependencies of nucleotides at different positions in RNA sequences,while ignoring the structural dependencies of nucleotides in their threedimensional *** overcome this issue,we propose a cross-species end-to-end deep learning model,namely CR-NSSD,which conduct a cross-domain representation learning process integrating nucleotide structural and sequential dependencies for RNA m6A site ***,CR-NSSD first obtains the pre-coded representations of RNA sequences by incorporating the position information into single-nucleotide states with chaos game representation *** then constructs a crossdomain reconstruction encoder to learn the sequential and structural dependencies between *** minimizing the reconstruction and binary cross-entropy losses,CR-NSSD is trained to complete the task of m6A site *** experiments have demonstrated the promising performance of CR-NSSD by comparing it with several state-of-the-art m6A identification ***,the results of cross-species prediction indicate that the integration of sequential and structural dependencies allows CR-NSSD to capture general features of m6A modification sites among different species,thus improving the accuracy of cross-species identification.

关键词： Cross-domain reconstruction cross-species prediction N6-methyladenosine(m6A)modification site RNA sequence sequential and structural dependencies

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Efficient Clustering Network Based on Matrix Factorization

引用

computers, Materials & Continua 2024年第7期80卷 281-298页

作者： Jieren Cheng Jimei li Faqiang zeng Zhicong Tao and Yue Yang School of computer science and Technology Hainan UniversityHaikou570228China School of Cyberspace Security Hainan UniversityHaikou570228China Hainan Blockchain Technology Engineering Research Center Haikou570228China

Contrastive learning is a significant research direction in the field of deep ***,existing data augmentation methods often lead to issues such as semantic drift in generated views while the complexity of model pre-training limits further improvement in the performance of existing *** address these challenges,we propose the Efficient Clustering Network based on Matrix Factorization(ECN-MF).Specifically,we design a batched low-rank Singular Value Decomposition(SVD)algorithm for data augmentation to eliminate redundant information and uncover major patterns of variation and key information in the ***,we design a Mutual Information-Enhanced Clustering Module(MI-ECM)to accelerate the training process by leveraging a simple architecture to bring samples from the same cluster closer while pushing samples from other clusters *** experiments on six datasets demonstrate that ECN-MF exhibits more effective performance compared to state-of-the-art algorithms.

关键词： Contrastive learning clustering matrix factorization

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Joint training with local soft attention and dual cross-neighbor label smoothing for unsupervised person re-identification

引用

Computational Visual Media 2024年第3期10卷 543-558页

作者： Qing Han Longfei li Weidong Min Qi Wang Qingpeng zeng Shimiao Cui Jiongjin Chen School of Mathematics and computer sciences Nanchang UniversityNanchang 330031China Institute of Metaverse Nanchang UniversityNanchang 330031China Jiangxi Key Laboratory of Smart City Nanchang 330031China

Existing unsupervised person re-identification approaches fail to fully capture thefine-grained features of local regions,which can result in people with similar appearances and different identities being assigned the same label after *** identity-independent information contained in different local regions leads to different levels of local *** address these challenges,joint training with local soft attention and dual cross-neighbor label smoothing(DCLS)is proposed in this ***,the joint training is divided into global and local parts,whereby a soft attention mechanism is proposed for the local branch to accurately capture the subtle differences in local regions,which improves the ability of the re-identification model in identifying a person’s local significant ***,DCLS is designed to progressively mitigate label noise in different local *** DCLS uses global and local similarity metrics to semantically align the global and local regions of the person and further determines the proximity association between local regions through the cross information of neighboring regions,thereby achieving label smoothing of the global and local regions throughout the training *** extensive experiments,the proposed method outperformed existing methods under unsupervised settings on several standard person re-identification datasets.

关键词： person re-identification(Re-ID) unsupervised learning(USL) local soft attention joint training dual cross-neighbor label smoothing(DCLS)

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：