A5101514218- Network Security and Intrusion Detection
- Anomaly Detection Techniques and Applications
- Video Surveillance and Tracking Methods
- Advanced Image and Video Retrieval Techniques
- Computational Fluid Dynamics and Aerodynamics
- Software System Performance and Reliability
- Fluid Dynamics and Turbulent Flows
- Ideological and Political Education
- Laser Material Processing Techniques
- Granular flow and fluidized beds
- Advanced Decision-Making Techniques
- Advanced Numerical Methods in Computational Mathematics
- Video Analysis and Summarization
- Surface Roughness and Optical Measurements
- Advanced machining processes and optimization
- Evaluation and Optimization Models
- Cyclone Separators and Fluid Dynamics
- Time Series Analysis and Forecasting
- Mineral Processing and Grinding
- China's Ethnic Minorities and Relations
- Geoscience and Mining Technology
Huawei Technologies (China)
2021-2022
Communication University of China
2011
Nanjing Institute of Technology
2011
Ling Tung University
2005
Beibu Gulf University
2002
Anomaly detection on multivariate time series (MTS) is an important research topic in data mining, which has a wide range of applications information technology, financial management, manufacturing system, and so on. However, the state-of-the-art unsupervised deep learning models for MTS anomaly are vulnerable to noise have poor performance training containing anomalies. In this article, we propose novel Self-Training based Detection with Generative Adversarial Network (GAN) model called...
In processing of laser cutting metal sheet, there is the complex correspondence between crafts parameter and surface quality. This paper proposed a control model quality based on Back-Propagetion(BP) neural networks, which indicated relationship craft parameters roughness cutting. By use simulated annealing, fitting accuracy convergence network would been improved. Experimental design with Central Composite Design could increase amount information reliability trainning samples. After...
This study is designed to achieve a video browsing technology which allows users browse the moving target of interest by controlling object in scene. included following: first, tracking accuracy needed be improved; second, we accelerate speed ensure that could view real-time changes object; third, user-friendly interactive interface was needed. In this study, applied both pyramid Lucas-Kanade optical flow algorithm and Camshift track target, optimized them using parallel processing...