A5072040420- Industrial Vision Systems and Defect Detection
- Image Processing Techniques and Applications
- Adaptive Dynamic Programming Control
- Iterative Learning Control Systems
- Mobile Agent-Based Network Management
- Reinforcement Learning in Robotics
- Viral Infections and Vectors
- Music and Audio Processing
- Advanced Image Processing Techniques
- Image and Object Detection Techniques
- Adaptive Control of Nonlinear Systems
- Speech and Audio Processing
- Surface Roughness and Optical Measurements
- Animal Vocal Communication and Behavior
Beijing University of Posts and Telecommunications
2023-2024
Xi'an Polytechnic University
2022-2023
The detection and location of yarn-dyed fabric defects is a crucial challenging problem in actual production scenarios. Recently, unsupervised defect methods based on convolutional neural networks have attracted more attention. However, the often neglect to model global receptive field images, which further influence ability model. In this article, we propose U-shaped Swin Transformer network Quadtree attention framework for detection. method via Transformer, adopts local effectively learn...
With the continuous promotion of “smart cities” worldwide, approach to be used in combining smart cities with modern advanced technologies (Internet Things, cloud computing, artificial intelligence) has become a hot topic. However, due non-stationary nature environmental sound and interference urban noise, it is challenging fully extract features from model single input achieve ideal classification results, even deep learning methods. To improve recognition accuracy ESC (environmental...
This study addresses the issue of slow speed and low efficiency in path planning algorithms for multi-agents on a plane. We formulate multi-agent problem as non-zero-sum stochastic game employ reinforcement learning algorithm, win or learn fast - policy hill-climbing (WoLF-PHC), to obtain Nash equilibrium strategy make conflict free optimal decisions each agent. Additionally, we propose adaptive WoLF-PHC (FA-WoLF-PHC) algorithm improve by constructing an objective function using gradient...
Automatic defect detection is an essential and challenging problem in the yarn-dyed weaving production process, this paper proposed a novel U-shaped Swin Transformer auto-encoder reconstructed model for shirt piece detection. This method uses of to extract global features image better reconstruct it more accurately, which solves problems scarce unbalanced type samples high cost actual production. Firstly, certain pattern, using defect-free adding Gaussian noise train reconstruction model....
Abstract To reduce the learning time and space occupation, this study presents a novel model‐free algorithm for obtaining Nash equilibrium solution of continuous‐time nonlinear non‐zero‐sum games. Based on integral reinforcement method, new HJ equation that can quickly cooperatively determine strategies all players is proposed. By leveraging neural network approximation gradient descent simultaneous adaptive tuning laws are provided both critic actor weights. These facilitate estimation...