- Architecture and Computational Design
- Urban Design and Spatial Analysis
- Healthcare and Venom Research
- Traditional Chinese Medicine Analysis
- Vehicle Dynamics and Control Systems
- Traditional Chinese Medicine Studies
- Advanced Data and IoT Technologies
- Computer Graphics and Visualization Techniques
- Islanding Detection in Power Systems
- Advanced Neural Network Applications
- Real-time simulation and control systems
- Remote-Sensing Image Classification
- Embedded Systems and FPGA Design
- Remote Sensing and Land Use
- Advanced Vision and Imaging
- Simulation and Modeling Applications
- Advanced Numerical Analysis Techniques
- Power Line Inspection Robots
- Urban Planning and Landscape Design
Sichuan University
2024
Wuhan University of Technology
2015
Chang'an University
2010
As a special insulation control, insulators protect transmission lines. In the process of detecting by UAV, locating is prerequisite for defect detection. However, complex line contexts, heavily occluded are abundant. This kind insulator difficult to locate accurately due its small size and extreme aspect ratio. To address this issue, an occlusion detection system based on YOLOX proposed in work. Firstly, improved SPP module used extract semantic information backbone's P5 layer. Secondly,...
As the fact that existing positive feedback active frequency drift (AFDPF) islanding detection algorithm has contradiction between time and total harmonic distortion (THD). In this paper, a variable coefficient is provided, which based on analysis of traditional method. The simulation model established MATLAB/SIMULINK, to testify function improved algorithm. It can be seen proposed method not only detected island effect effectively, but also had less blind detection.
A classification problem involving multi-class samples is typically divided into a set of two-class sub-problems. The pairwise probabilities produced by the binary classifiers are subsequently combined to generate final result. However, only that have been trained with unknown real class an unlabeled sample relevant problem. distance-based relative competence weighting (DRCW) combination mechanism can estimate classifiers. In this work, we adapt DRCW support vector machine (SVM) approach for...
The automobile power parameters need to be estimated for consumers, even if the measured data by professional devices does not exist. Using a little data, fitting or optimal switching methods still can support important approximately, such as power, speed, acceleration and so on. A least time switched computation of is put forward process also demonstrated in detail. For family used car examples, performance are computed out Mat lab programs. whole route method given movement rule, which...
This paper presents a vehicle driving simulator system with traffic oriented visual modeler and novel human-computer interaction way. Especially, the is based on mixed reality technology which equipped webcam printed controlling markers. Then used as virtual wheel to control motion of vehicle. At such low cost, many people can enjoy driving; it also provides an easy interesting way for video games promising business values.