A5100345147- Quantum Chromodynamics and Particle Interactions
- Particle physics theoretical and experimental studies
- Advanced Antenna and Metasurface Technologies
- Remote Sensing and Land Use
- Antenna Design and Analysis
- High-Energy Particle Collisions Research
- Geology and Paleoclimatology Research
- Groundwater and Isotope Geochemistry
- High-Temperature Coating Behaviors
- Coal and Its By-products
- Metal and Thin Film Mechanics
- Metamaterials and Metasurfaces Applications
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Geochemistry and Elemental Analysis
- Marine and coastal plant biology
- Paleontology and Stratigraphy of Fossils
- Adversarial Robustness in Machine Learning
- Advanced materials and composites
- Advanced SAR Imaging Techniques
- Medical Imaging Techniques and Applications
- Anomaly Detection Techniques and Applications
- Full-Duplex Wireless Communications
- Remote-Sensing Image Classification
- Advanced ceramic materials synthesis
- Microwave Engineering and Waveguides
National University of Defense Technology
2019-2025
University of Science and Technology of China
2019-2024
Hefei University of Technology
2022
Sun Yat-sen University
2022
Xiangtan University
2021
NARI Group (China)
2020
Shenyang Aerospace University
2017
China University of Geosciences (Beijing)
2017
Central China Normal University
2011
Land cover classification using high-resolution Polarimetric Synthetic Aperture Radar (PolSAR) images obtained from satellites is a challenging task. While deep learning algorithms have been extensively studied for PolSAR image land classification, the performance severely constrained due to scarcity of labeled samples and limited domain acceptance models. Recently, emergence Segment Anything Model (SAM) based on vision transformer (VIT) model has brought about revolution in study specific...
The threat posed by adversarial examples (AEs) to deep learning applications has garnered significant attention from the academic community. In response, various defense strategies have been proposed, including example detection. A range of detection algorithms developed differentiate between benign samples and examples. However, accuracy these is significantly influenced characteristics attacks, such as attack type intensity. Furthermore, impact image preprocessing on robustness—a common...
The rapid proliferation of Internet Things (IoT) devices has led to an increased need for robust and efficient intrusion detection systems capable identifying mitigating novel threats. Traditional methods often struggle with the scarcity labeled anomaly data, which is highly consequential, particularly in context IoT. In this study, we propose a few-shot learning approach by leveraging Multi-Stage Attention Siamese Network (MASiNet) network traffic based on meta-learning framework. Unlike...
Optical and Synthetic Aperture Radar (SAR) imagery offers a wealth of complementary information on given target, attributable to the distinct imaging modalities each component image type. Thus, multimodal remote sensing data have been widely used improve land cover classification. However, fully integrating optical SAR is not straightforward due distributions their features. To this end, we propose classification network based feature fusion, i.e., MFFnet. We adopt dual-stream extract...
With the continuous progress of remote sensing technology, a large number airborne and satellite-based polarimetric SAR have been applied in field to provide decision basis for military intelligence, land use, urban planning. Terrain classification PolSAR image has become an important research direction interpretation. In this paper, data Flevoland area Netherlands are selected. First, coherent speckle noise is suppressed by Refined Lee filtering, 68-dimensional polarization features...
Abstract Miniaturized single‐/dual‐band bandpass filters (BPFs) are proposed in this paper based on one single grounded square patch (GSP) resonator. The GSP resonator is realized by introducing ring‐shaped metallic vias at the center of a patch. Owing to specific geometry, an additional mode nominated as LC generated, which makes operate below fundamental resonate frequency non‐grounded patch, thus miniaturization has been achieved. In addition, several slots etched serve perturbations and...
The complex and variable ocean environment noise has a very bad influence on the sonar detection signal, which leads to low probability of finding target. adaptive filter can uncertain track change continuously adjust parameters achieve optimal filtering effect. As traditional algorithm disadvantages convergence speed steady-state error, poor effect impulse noise, an step-size LMS is proposed according characteristics underwater sources. results show that converge quickly, maintain small...
Metalens can achieve arbitrary light modulation by controlling the amplitude, phase, and polarization of incident waves have been applied across various fields. This paper presents a color router designed based on metalens, capable effectively separating spectra from visible to near-infrared light. Traditional design methods for meta-lenses require extensive simulations, making them time-consuming. In this study, we propose deep learning network forward prediction broad wavelength range,...
System size is more than a geometrical quantity in relativistic heavy ion collisions; it closely related to evolution process, i.e. different system corresponds and whether QGP produced depends on the size. We propose that should be under same level when comparing measurements from colliding nuclei. The equivalence of peripheral collisions Au-Au central smaller nuclei studied using Monte Carlo method. Comparing transverse overlapping area nuclei, number participant nucleons nucleon-nucleon...
A wideband hybrid decoupling structure for MIMO antennas is presented in this paper. The composed of a suspended array-antenna surface, which consists several reflectors and defected ground with two rectangular slots to reduce the mutual coupling. When utilizing proposed design extremely-closed patch an edge-to-edge distance only 1 mm (0.019 λ <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</inf> at 5.65 GHz), simulated results demonstrate...