A5108889353- Emotion and Mood Recognition
- Face and Expression Recognition
- Lightning and Electromagnetic Phenomena
- Earthquake Detection and Analysis
- Face recognition and analysis
- Anomaly Detection Techniques and Applications
- Satellite Communication Systems
- Advanced Computing and Algorithms
- Mobile Agent-Based Network Management
- Distributed and Parallel Computing Systems
- Seismic Waves and Analysis
- Advanced Malware Detection Techniques
- Network Security and Intrusion Detection
- Ionosphere and magnetosphere dynamics
Technology and Engineering Center for Space Utilization
2017-2024
Chinese Academy of Sciences
2019-2024
University of Chinese Academy of Sciences
2022
Facial Expression Recognition (FER) in the wild is an extremely challenging task. Recently, some Vision Transformers (ViT) have been explored for FER, but most of them perform inferiorly compared to Convolutional Neural Networks (CNN). This mainly because new proposed modules are difficult converge well from scratch due lacking inductive bias and easy focus on occlusion noisy areas. TransFER, a representative transformer-based method alleviates this with multi-branch attention dropping...
Facial expression recognition plays an important role in human-computer interaction. In this paper, we propose the Coarse-to-Fine Cascaded network with Smooth Predicting (CFC-SP) to improve performance of facial recognition. CFC-SP contains two core components, namely networks (CFC) and (SP). For CFC, it first groups several similar emotions form a rough category, then employs conduct coarse but accurate classification. Later, additional for these grouped is further used obtain fine-grained...
Geomagnetic storms have a significant impact on Earth’s magnetosphere and ionosphere, as well the global atmospheric circuit. This study focuses investigating anomalous variations in vertical electric field at eight mid-latitude low-latitude stations during mega-geomagnetic storm 24 April 2023. The majority of observed increases, while only three nearby exhibited decreases. analysis revealed that changes ranged from 19 to 370 V/m, time differences between extreme values minimum Dst value 0...
In fair weather, the vertical atmospheric electric field is oriented downward (positive in earth surface ordinate system) global circuit. Some researchers have revealed unique phenomenon whereby once an upward observed earthquake (EQ) follows within 2–48 h regardless of EQ magnitude. However, mechanism has not been explained with a suitable physical model. this paper, model presented considering four types forces acting on charged particles air. It demonstrated that heavier positive ions and...
The backbone network of the Space Information Network (SIN) contains high orbit satellites, medium satellites and low earth satellites. fault detection diagnosis capability is key to healthy operation SIN. In this paper, we present a test model for network, adopting way clustering classify different clusters convenience manage network. Specially, each cluster has fixed topology. It very handy detection. We exploit FTA (fault tree analysis) analyze SIN from testing satellite And by analysis,...
Firstly, we introduce intrusion detection system and anomaly detection. And then do some research on machine learning techniques for by network dataset NSL-KDD. The algorithms such as J48, Random forest, SVM, Vote, Stacking are selected. Forest, Vote stacking ensemble methods. We try to test verify performance of multiple methods a 20 per cent NSL-KDD experiment. experiment data has two parts. First, the is classified into normal anomaly. Second, feature attack type added dataset, new...
Facial expression recognition plays an important role in human-computer interaction. In this paper, we propose the Coarse-to-Fine Cascaded network with Smooth Predicting (CFC-SP) to improve performance of facial recognition. CFC-SP contains two core components, namely networks (CFC) and (SP). For CFC, it first groups several similar emotions form a rough category, then employs conduct coarse but accurate classification. Later, additional for these grouped is further used obtain fine-grained...