检索结果-南通市图书馆

Logit prototype learning with active multimodal representation for robust open-set recognition

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第6期67卷 297-312页

作者： Yimin fu Zhunga liu Zicheng WANG School of Automation Northwestern Polytechnical University

Robust open-set recognition(OSR) performance has become a prerequisite for pattern recognition systems in real-world applications. However, the existing OSR methods are primarily implemented on the basis of single-modal perception, and their performance is limited when single-modal data fail to provide sufficient descriptions of the objects. Although multimodal data can provide more comprehensive information than single-modal data, the learning of decision boundaries can be affected by the feature representation gap between different modalities. To effectively integrate multimodal data for robust OSR performance, we propose logit prototype learning(LPL) with active multimodal representation. In LPL, the input multimodal data are transformed into the logit space, enabling a direct exploration of intermodal correlations without the impact of scale inconsistency. Then, the fusion weights of each modality are determined using an entropybased uncertainty estimation method. This approach realizes adaptive adjustment of the fusion strategy to provide comprehensive descriptions in the presence of external disturbances. Moreover, the single-modal and multimodal representations are jointly optimized interactively to learn discriminative decision ***, a stepwise recognition rule is employed to reduce the misclassification risk and facilitate the distinction between known and unknown classes. Extensive experiments on three multimodal datasets have been done to demonstrate the effectiveness of the proposed method.

关键词： logit prototype learning multimodal perception open-set recognition uncertainty estimation

LMR-CBT: learning modality-fused representations with CB-Transformer for multimodal emotion recognition from unaligned multimodal sequences

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2024年第4期18卷 39-47页

作者： Ziwang fu Feng liu Qing XU Xiangling fu Jiayin QI School of Computer Science(National Pilot Software Engineering School) Beijing University of Posts and TelecommunicationsBeijing 100876China Key Laboratory of Trustworthy Distributed Computing and Service(BUPT) Beijing 100876China Shanghai International School of Chief Technology Officer East China Normal UniversityShanghai 200062China School of Computer Science and Technology East China Normal UniversityShanghai 200062China School of Cyberspace Security Guangzhou UniversityGuangdong 510006China

Learning modality-fused representations and processing unaligned multimodal sequences are meaningful and challenging in multimodal emotion *** approaches use directional pairwise attention or a message hub to fuse language,visual,and audio ***,these fusion methods are often quadratic in complexity with respect to the modal sequence length,bring redundant information and are not *** this paper,we propose an efficient neural network to learn modality-fused representations with CB-Transformer(LMR-CBT)for multimodal emotion recognition from unaligned multi-modal ***,we first perform feature extraction for the three modalities respectively to obtain the local structure of the ***,we design an innovative asymmetric transformer with cross-modal blocks(CB-Transformer)that enables complementary learning of different modalities,mainly divided into local temporal learning,cross-modal feature fusion and global self-attention *** addition,we splice the fused features with the original features to classify the emotions of the ***,we conduct word-aligned and unaligned experiments on three challenging datasets,IEMOCAP,CMU-MOSI,and *** experimental results show the superiority and efficiency of our proposed method in both *** with the mainstream methods,our approach reaches the state-of-the-art with a minimum number of parameters.

关键词： modality-fused representations cross-model blocks multimodal emotion recognition unaligned multimodal sequences computational affection

Vanadium nitride quantum dots@carbon skeleton anode material synthesized via in situ oxidation initiation strategy

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Tungsten 2024年第3期6卷 561-573页

作者： Yi-Han fu Yuan-you Peng Lei Zhao Tian-Qi He Mei-Mei Yuan Hao Dang Rui liu Fen Ran State Key Laboratory of Advanced Processing and Recycling of Non-Ferrous Metals School of Materials Science and Engineering Lanzhou University of Technology

Vanadium nitride（VN） has received significant attention as a potential negative electrode material due to its remarkable theoretical and physical ***,the electrochemical performance needs to be improved due to its low specific surface area and the issue of being easily dissolved in *** have shown that the composite structure of carbon-coated nano sc ale VN particles could significantly increase the specific surface area of VN,and more importantly,and the carbon matrix can effectively restrain the dissolution of VN particles in the electrolyte,thereby improving its chemical and electrochemical performance,including specific capacity and cycling ***,VN quantum dots are synthesized and protected by uniformly coating the carbon skeleton on its surface via an in situ strategy of oxidation initiation,which initiates dopamine polymerization,forming a crystalline complex of polydopamine and *** heat treatment,the product is a composite material with VN quantum dots embedded in the amorphous carbon *** resulting smallsized and crosslinked carbon particles formed a graded pore structure,more reasonable channels provided for electrolyte ions transport *** the composite material was used as the negative electrode material for supercapacitors,it exhibits the specific capacity of 391.9 F·g-1at a current density of 0.5 A·g-1and excellent capacity retention rate of 83.6% after *** synthesis method of this quantum dot could be applied to other metal compounds and has potential applications in catalysis,environmental protection,and other fields.

关键词： In situ initiation Vanadium nitride Quantum dots Anode Supercapacitors

Low-Complexity Reconstruction of Covariance Matrix in Hybrid Uniform Circular Array

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

China Communications 2024年第3期21卷 66-74页

作者： fu Zihao liu Yinsheng Duan Hongtao Information School Communication University of ChinaBeijing 100024China Beijing Institute of Electronic System Engineering Beijing 100854China School of Electronic and Information Engineering Beijing Jiaotong UniversityBeijing 100044China Beijing Radio Monitoring Station of State Radio Monitoring Center(SRMC) Beijing 100037China

Spatial covariance matrix(SCM) is essential in many multi-antenna systems such as massive multiple-input multiple-output(MIMO). For multi-antenna systems operating at millimeter-wave bands, hybrid analog-digital structure has been widely adopted to reduce the cost of radio frequency *** this situation, signals received at the antennas are unavailable to the digital receiver, and as a consequence, traditional sample average approach cannot be used for SCM reconstruction in hybrid multi-antenna systems. To address this issue, beam sweeping algorithm(BSA) which can reconstruct the SCM effectively for a hybrid uniform linear array, has been proposed in our previous works. However, direct extension of BSA to a hybrid uniform circular array(UCA)will result in a huge computational burden. To this end, a low-complexity approach is proposed in this paper. By exploiting the symmetry features of SCM for the UCA, the number of unknowns can be reduced significantly and thus the complexity of reconstruction can be saved accordingly. furthermore, an insightful analysis is also presented in this paper, showing that the reduction of the number of unknowns can also improve the accuracy of the reconstructed SCM. Simulation results are also shown to demonstrate the proposed approach.

关键词： hybrid array millimeter-wave spatial covariance matrix uniform circular array

Super-resolution Imaging of Telescopic Systems based on Optical-neural Network Joint Optimization

在线全文

学校读者我要写书评

暂无评论

Research in Astronomy and Astrophysics 2024年第9期24卷 167-177页

作者： you-Hong Sun Tao Zhang Hao-Dong Shi Qiang fu Jia-Nan liu Kai-Kai Wang Chao Wang Research Institute of Chongqing Changchun University of Science and TechnologyChangchun 130022China National and Local Joint Engineering Research Center for Space Photoelectric Technology Changchun University of Science and TechnologyChangchun 130022China The Institute of Remote Sensing Satellites China Academy of Space Technology(CAST)Beijing 100094China

Optical telescopes are an important tool for acquiring optical information about distant objects,and resolution is an important indicator that measures the ability to observe object ***,due to the effects of system aberration,atmospheric seeing,and other factors,the observed image of ground-based telescopes is often degraded,resulting in reduced *** paper proposes an optical-neural network joint optimization method to improve the resolution of the observed image by co-optimizing the point-spread function(PSF)of the telescopic system and the image super-resolution(SR)*** improve the speed of image reconstruction,we designed a generative adversarial net(LCR-GAN)with light parameters,which is much faster than the latest unsupervised *** reconstruct the PSF trained by the network in the optical path,a phase mask is *** improves the image reconstruction effect of LCR-GAN by reconstructing the PSF that best matches the *** results of simulation and verification experiments show that compared with the pure deep learning method,the SR image reconstructed by this method is rich in detail and it is easier to distinguish stars or stripes.

关键词： Techniques:image processing Telescopes Stars:imaging

Gyroid Triply Periodic Minimal Surface Lattice Structure Enables Improved Superelasticity of CuAlMn Shape Memory Alloy

在线全文

学校读者我要写书评

暂无评论

Acta Metallurgica Sinica(English Letters) 2024年第6期37卷 1047-1065页

作者： Mengwei Wu Chunmei Ma Ruiping liu Huadong fu Department of Materials Science and Engineering China University of Mining and Technology(Beijing)Beijing100083China Institute for Advanced Materials and Technology University of Science and Technology BeijingBeijing100083China

Improving the shape memory effect and superelasticity of Cu-based shape memory alloys(SMAs)has always been a research hotspot in many *** work systematically investigates the effects of Gyroid triply periodic minimal surface(TPMS)lattice structures with different unit sizes and volume fractions on the manufacturing viability,compressive mechanical response,superelasticity and heating recovery properties of CuAlMn *** results show that the increased specific surface area of the lattice structure leads to increased powder adhesion,making the manufacturability proportional to the unit size and volume *** compressive response of the CuAlMn SMAs Gyroid TPMS lattice structure is negatively correlated with the unit size and positively correlated with the volume *** superelastic recovery of all CuAlMn SMAs with Gyroid TPMS lattice structures is within 5%when the cyclic cumulative strain is set to be 10%.The lattice structure shows the maximum superelasticity when the unit size is 3.00 mm and the volume fraction is 12%,and after heating recovery,the total recovery strain increases as the volume fraction *** study introduces a new strategy to enhance the superelastic properties and expand the applications of CuAlMn SMAs in soft robotics,medical equipment,aerospace and other fields.

关键词： Shape memory alloys Superelasticity Gyroid triply periodic minimal surface(TPMS)lattice structure Selective laser melting(SLM)

Solvent transport dynamics and its effect on evolution of mechanical properties of nitrocellulose(NC)-based propellants under hot-air drying process

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Defence Technology（防务技术） 2024年第2期32卷 262-270页

作者： Enfa fu Mingjun Yi Qianling liu Zhenggang Xiao Key Laboratory of Special Energy Materials Ministry of EducationSchool of Chemistry and Chemical EngineeringNanjing University of Science and TechnologyNanjing 210094People's Republic of China

Appropriate drying process with optimized controlling of drying parameters plays a vital role in the improvement of the quality and performance of propellant ***,few research on solvent transport dynamics within NC-based propellants was reported,and its effect on the evolution of mechanical properties was not interpreted *** study is conducted to gain a comprehensive understanding of hot-air drying for NC-based propellants and clarify the effect of temperature on solvent transport behavior and further the change of mechanical properties during *** drying kinetic curves show the drying time required is decreased but the steady solvent content is increased and the drying rate is obviously increased with the increase of hot-air temperatures,indicating hot-air temperatures have a significant effect on drying kinetics.A modified drying model was established,and results show it is more appropriate to describe solvent transport behavior within NC-based ***,two linear equations were established to exhibit the relationship between solvent content and its effect on the change of tensile properties,and the decrease of residual solvent content causes an obvious increase of tensile strength and tensile modulus of propellant products,indicating its mechanical properties can be partly improved by adjustment of residual solvent *** outcomes can be used to clarify solvent transport mechanisms and optimize drying process parameters of double-based gun propellants.

关键词： Nitrocellulose-based propellants Solvent transport dynamics Mechanical properties Drying kinetics Effective solvent diffusion coefficient

Recent advances in morphology control of platinum catalysts toward oxygen reduction reaction

在线全文

学校读者我要写书评

暂无评论

Frontiers in Energy 2024年第3期18卷 330-355页

作者： Shun CHEN Yanru liu Xiaogang fu Wanglei WANG State Key Laboratory of Solidification Processing Atomic Control and Catalysis Engineering LaboratorySchool of Materials Science and EngineeringNorthwestern Polytechnical UniversityXi’an 710072China State Key Laboratory of Oral and Maxillofacial Reconstruction and Regeneration National Clinical Research Center for Oral DiseasesShaanxi Key Laboratory of StomatologyDepartment of Dental MaterialsSchool of StomatologyThe Fourth Military Medical UniversityXi’an 710032China

Exploring advanced platinum (Pt)-based electrocatalysts is vital for the widespread implementation of proton exchange membrane fuel cells (PEMFCs). Morphology control represents an effective strategy to optimize the behavior of Pt catalysts. In this work, an attempt is made to comprehensively review the effect of morphology control on the catalytic behavior of catalysts in the oxygen reduction reaction (ORR). First, the fundamental physicochemical changes behind morphology control, including exposing more active sites, generating appropriate lattice strains, and forming different crystalline surfaces, are highlighted. Then, recently developed strategies for tuning the morphologies of electrocatalysts, including core-shell structures, hollow structures, nanocages, nanowires, and nanosheets, are comprehensively summarized. Finally, an outlook on the future development of morphology control of Pt catalysts is presented, including rational design strategies, advanced in situ characterization techniques, novel artificial intelligence, and mechanical learning. This work is intended to provide valuable insights into designing the morphology and technological innovation of efficient redox electrocatalysts in fuel cells.

关键词： morphology platinum catalysts electrocatalysis ORR PEMFC

Tripartite motif-containing protein 11(TRIM11):a novel weapon against Alzheimer’s disease

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Signal Transduction and Targeted Therapy 2024年第2期9卷 339-340页

作者： Yichu fu Bi Zhang Jianfeng liu College of Life Science and Technology Key Laboratory of Molecular Biophysics of MOEHuazhong University of Science and TechnologyWuhanHubei 430074China

In a recent study published in Science,Zhang et *** that tripartite motif-containing(TRIM)protein 11(TRIM11)exhibited strong inhibition on tau aggregation,which is a pathological characteristic of tauopathies including Alzheimer’s disease(AD).1 This makes TRIM11 a promising novel target for the treatment of AD and other tauopathies.

关键词： Alzheimer tau TRIM