A5032664588- Advanced Neural Network Applications
- Catalytic Processes in Materials Science
- Electrochemical sensors and biosensors
- Reinforcement Learning in Robotics
- Indoor and Outdoor Localization Technologies
- Perovskite Materials and Applications
- Electrochemical Analysis and Applications
- Occupational Health and Safety Research
- Brain Tumor Detection and Classification
- Risk and Safety Analysis
- Catalysts for Methane Reforming
- Domain Adaptation and Few-Shot Learning
- Human-Automation Interaction and Safety
- Air Quality Monitoring and Forecasting
- Advanced Image and Video Retrieval Techniques
- Medical Image Segmentation Techniques
- Electrocatalysts for Energy Conversion
- Robotics and Sensor-Based Localization
- Advanced Vision and Imaging
- Advanced Image Processing Techniques
- Advancements in Solid Oxide Fuel Cells
- Conducting polymers and applications
- 2D Materials and Applications
- Machine Fault Diagnosis Techniques
- Railway Engineering and Dynamics
Tianjin Medical University
2024-2025
Tianjin University
2024-2025
Shanghai Jiao Tong University
2025
University of Science and Technology Liaoning
2023-2024
Jiangxi Normal University
2023-2024
Harbin Engineering University
2022-2024
Shandong University of Technology
2021-2024
East China University of Science and Technology
2024
Chongqing Jiaotong University
2024
Tsinghua University
2019-2023
In recent years, wireless sensing has been exploited as a promising research direction for contactless human activity recognition. However, one major issue hindering the real deployment of these systems is that signal variation patterns induced by activities with different devices and environmental settings are neither stable nor consistent, resulting in unstable system performance. The existing machine learning based methods usually take "black box" approach fails to achieve consistent this...
Recent years have witnessed advances of Internet Things (IoT) technologies and their applications to enable contactless sensing human-computer interaction in smart homes. For people with motor neurone disease (MND), motion capabilities are severely impaired they difficulties interacting IoT devices even communicating other people. As the progresses, most patients lose speech function eventually which makes widely adopted voice-based solutions fail. In contrast, can still move fingers...
Designing accurate and efficient convolutional neural architectures for vast amount of hardware is challenging because designs are complex diverse. This paper addresses the diversity challenge in Neural Architecture Search (NAS). Unlike previous approaches that apply search algorithms on a small, human-designed space without considering diversity, we propose HURRICANE explores automatic hardware-aware over much larger two-stage algorithm, to efficiently generate tailored models different...
Background: As urbanization progresses, the resulting air pollution has become an increasingly severe public health issue, known to exacerbate respiratory and cardiovascular diseases. Recent studies suggest that it may also affect blood coagulation mechanisms. In this study, Mendelian randomization (MR) analysis was used explore causal relationship between pulmonary embolism (PE). Methods: This study employs MR techniques, using genetic variants associated with exposure assess their impact...
The scaling of advanced integrated circuits has posed significant challenges for traditional Cu interconnects, including increased resistivity and reduced electromigration lifetime. Materials with high cohesive energy low ρ0 × λ values are emerging as promising alternatives. In this work, active learning coupling density functional theory (DFT) computation is employed to accelerate the discovery binary intermetallic compounds interconnect materials. Following five iterations, 100 screened...
6D pose estimation from a single RGB image is challenging and vital task in computer vision. The current mainstream deep model methods resort to 2D images annotated with real-world ground-truth object poses, whose collection fairly cumbersome expensive, even unavailable many cases. In this work, get rid of the burden annotations, we formulate refinement as Markov Decision Process impose on reinforcement learning approach only annotations weakly-supervised information, via delicate reward...
Abstract Circularly polarized light (CPL) detection has wide applications in many fields, where the anisotropy factor ( g Iph ) is an important indicator to characterize CPL performance. So far, materials with high have been reported, however, exploration of regulation still its infancy. Herein, two novel alternating chiral‐achiral cations intercalation‐type chiral hybrid perovskites (CHPs), named R / S ‐1‐phenylpropylamine)(propylamine)PbBr 4 (1‐ ), exhibit above room‐temperature (RT)...
Metal halide perovskite ferroelectrics combining spontaneous polarization and excellent semiconducting properties is an ideal platform for enabling self-driven X-ray detection. However, achievements to date have been only based on uniaxiality, which increases the complexity of device fabrication. Multi-axial ferroelectric materials multiple equivalent directions, making them potentially amenable multi-axial detection, but report these types still a huge blank. Herein, high-quality (BA)
The long-span dual-purpose highway-rail double-tower cable-stayed bridge has the characteristics of a large span and load-bearing capacity. Compared with traditional bridge, its wind resistance seismic are weaker, dynamic closely related to bearing capacity bridge. This study investigated influence variations member parameters on then improved To provide necessary experimental theory for research work bridges, this paper takes world’s longest as background, using finite element analysis...
In recent years, the rapid development of Deep Learning (DL) has provided a new method for ship detection in Synthetic Aperture Radar (SAR) images. However, there are still four challenges this task. (1) The targets SAR images very sparse. A large number unnecessary anchor boxes may be generated on feature map when using traditional anchor-based models, which could greatly increase amount computation and make it difficult to achieve real-time detection. (2) size is relatively small. Most...
The study of cloud types is critical for understanding atmospheric motions and climate predictions; example, accurately classified products help improve meteorological predicting accuracies. However, the current satellite classification methods generally analyze threshold change in a single pixel do not consider relationship between surrounding pixels. development relies heavily on human recourses does fully utilize data-driven advantages computer models. Here, new intelligent method based...
Abstract Copper (Cu) is a promising metal for electrochemical CO 2 reduction reaction (eCO RR) to value‐added C products. However, as the key intermediate C─C coupling form products, * difficult adsorb on Cu surface moderate level. Rare earth elements possess distinctive electronic structure that effectively regulates local density of surrounding atoms, yet has rarely been investigated. Herein, different rare metals are screened doping O and found Sc atomically dispersed can weaken...
Methane steam reforming (MSR) reaction is a promising industrial hydrogen production technology. In this study, the impregnation method was deployed to prepare mono Co and Co-based bimetallic catalysts loaded on CeO2 nanoparticles as for MSR. Results show that improved have good performance compared one. The best catalyst sample Co-Ni/CeO2 which resulted in CH4 conversion of 76.1%, H2 selectivity 58.5% yield 44.5%. Furthermore, increased significantly from 77.2% with increase total Co-Ni...
We report here, a strategy to prepare Pt/Pd nanoparticles decorated with Co-N-C materials, where was obtained via pyrolysis of ZIF-67 directly. As-prepared Pt/Pd/Co-N-C catalysts showed excellent ORR performance, offered higher limit current density (6.6 mA cm−2) and similar half-wave potential positive (E1/2 = 0.84 V) compared commercial Pt/C. In addition an activity, it also exhibits robust durability. The decreased by only 9% after adding methanol, 10% loss continuous testing at 36,000 s.
The Automated guided vehicle (AGV) has been one of the most popular topics for last few decades, on account industrial need higher efficiencies. Four basic functions are required an AGV system: mapping, navigation, dynamic collision avoidance, and coordination. commonly applied system is 2D Simultaneous Localization Mapping (2D SLAM) which utilizes a LiDAR at relatively low cost high efficiency. However, SLAM algorithm critical defect. It can only acquire information, leading to some...