A5100827628- Advanced Vision and Imaging
- Robotics and Sensor-Based Localization
- Image and Object Detection Techniques
- Radiomics and Machine Learning in Medical Imaging
- Soil Carbon and Nitrogen Dynamics
- Underwater Vehicles and Communication Systems
- Modular Robots and Swarm Intelligence
- EEG and Brain-Computer Interfaces
- Medical Image Segmentation Techniques
- Gaze Tracking and Assistive Technology
- Cell Image Analysis Techniques
- Surface Modification and Superhydrophobicity
- Fluid Dynamics and Heat Transfer
- Plant Water Relations and Carbon Dynamics
- Venomous Animal Envenomation and Studies
- Radar Systems and Signal Processing
- Advanced X-ray and CT Imaging
- Digital Imaging for Blood Diseases
- Graphene and Nanomaterials Applications
- Rabies epidemiology and control
- Advanced SAR Imaging Techniques
- Advanced Neural Network Applications
- Biomimetic flight and propulsion mechanisms
- Porphyrin and Phthalocyanine Chemistry
- Peatlands and Wetlands Ecology
Huazhong University of Science and Technology
2023-2025
Zhejiang Sci-Tech University
2024
Chinese Academy of Sciences
2002-2023
Xi'an Technological University
2023
State Key Laboratory of Magnetism
2023
Wuhan Institute of Physics and Mathematics
2023
Institute of Plasma Physics
2022
Fujian Institute of Research on the Structure of Matter
2022
University of Chinese Academy of Sciences
2022
Institute of Applied Ecology
2016
Segment anything model (SAM), an eminent universal image segmentation model, has recently gathered considerable attention within the domain of medical segmentation. Despite remarkable performance SAM on natural images, it grapples with significant degradation and limited generalization when confronted particularly those involving objects low contrast, faint boundaries, intricate shapes, diminutive sizes. In this paper, we propose SAMUS, a tailored for ultrasound contrast to previous...
Open AccessCCS ChemistryRESEARCH ARTICLE14 Feb 2022Achieving T-Type Photochromism through Generating Copper(I) Metallacyclopentadiene Biradical Xu Zhang, Jin-Yun Wang, Liang-Jin Xu, Xu-Yuan Jin, Xin Yang, Li-Yi Zhang and Zhong-Ning Chen State Key Laboratory of Structural Chemistry, Fujian Institute Research on the Structure Matter, Chinese Academy Sciences, Fuzhou, 350002 , Wang *Corresponding authors: E-mail Address: [email protected] Science Technology Innovation for Optoelectronic...
Recent approaches to VO have significantly improved performance by using deep networks predict optical flow between video frames. However, existing methods still suffer from noisy and inconsistent matching, making it difficult handle challenging scenarios long-sequence estimation.To overcome these challenges, we introduce Spatio-Temporal Visual Odometry (STVO), a novel network architecture that effectively leverages inherent spatio-temporal cues enhance the accuracy consistency of...
Stereo matching is a core component in many computer vision and robotics systems. Despite significant advances over the last decade, handling ambiguities ill-posed regions large disparities remains an open challenge. In this paper, we propose new deep network architecture, called IGEV++, for stereo matching. The proposed IGEV++ constructs Multi-range Geometry Encoding Volumes (MGEV), which encode coarse-grained geometry information disparities, while preserving fine-grained details small...
Inspired by the agility of mantis shrimp, this paper presents development a novel bionic shrimp robot with multiple pleopods coupled motion for underwater target recognition and tracking. Firstly, is constructed based on inspiration from biological shrimp. Secondly, Kalman filter-based algorithm, namely MobileNet-YOLO (KMY), created to collect datasets train neural networks. Thirdly, moving following system developed robot, in which dual PID controller used track targets. The real testing...
We investigated the fire impacts on nutrients in litter and soil, their C:N:P stoichio-metry forests of Great Xing'an Mountains. The studied sites differed burning year (post-fire 4, 14, 40, 70 years unburned within 120 years) had different topographic locations (sloped land flat land). results showed that there were significant differences characteristics C, N, P for both soil with years. No fluctuation was observed C content, while contents N increased increasing post-fire recovery In...
Rabies-viruses-based retrograde tracers can spread across multiple synapses in a direction the nervous system of rodents and primates, making them powerful tools for determining structure function complicated neural circuits brain. However, they have some limitations, such as posing high risks to human health inability trans-synaptic label inputs from genetically-defined starter neurons. Here, we established new trans-multi-synaptic tracing method through brain-wide rabies virus glycoprotein...
Spray cooling is an effective solution for high heat flux dissipation challenges. Accurate prediction of transfer efficiency by numerical simulation can reduce the cost spray in engineering applications. To improve accuracy simulation, this study develops a mathematical model droplet collision and response based on experimental data. In experiments, droplets are sprayed onto 200 °C aluminum alloy thermal wall using atomizing nozzle, temperature monitored, curve variation during estimated...
Recent approaches to VO have significantly improved performance by using deep networks predict optical flow between video frames. However, existing methods still suffer from noisy and inconsistent matching, making it difficult handle challenging scenarios long-sequence estimation. To overcome these challenges, we introduce Spatio-Temporal Visual Odometry (STVO), a novel network architecture that effectively leverages inherent spatio-temporal cues enhance the accuracy consistency of...
Visual odometry (VO) aims to estimate camera poses from visual inputs -- a fundamental building block for many applications such as VR/AR and robotics. This work focuses on monocular RGB VO where the input is video without IMU or 3D sensors. Existing approaches lack robustness under this challenging scenario fail generalize unseen data (especially outdoors); they also cannot recover metric-scale poses. We propose Robust Metric Odometry (RoMeO), novel method that resolves these issues...
A radar target detection system based on chaotic shape-forming filter is proposed in this paper. The design scheme uses the shaping signal as waveform, and ideal "thumbtack-shaped" ambiguity function of signal, low side lobe, narrow main lobe concentrated energy to improve range resolution system. sequence information predefined was used reference achieve high-fidelity storage reduce cost. Then, received were fed into corresponding matched filters signal-to-noise ratio receiver. Finally,...
Abstract Background: Reendothelialization is a natural pathway to inhibit neointimal hyperplasia and in-stent restenosis. Circulating endothelial progenitor cells (EPCs) derived from bone marrow might contribute repair. However, the temporal spatial distribution of injured vascular reendothelialization in presence transplanted EPCs not quite clear. Methods: We performed carotid balloon injury model treated by transplantation marrow-derived EPCs. To evaluate after injury, arteries were...