A fault diagnosis method for few-shot industrial processes based on semantic segmentation and hybrid domain transfer learning.
Ying TianYiwei WangXin PengWei ZhangPublished in: Appl. Intell. (2023)
Keyphrases
- transfer learning
- industrial processes
- semantic segmentation
- cross domain
- fault detection
- labeled data
- reinforcement learning
- target domain
- superpixels
- transfer knowledge
- machine learning
- active learning
- quality improvement
- semi supervised learning
- conditional random fields
- collaborative filtering
- weakly supervised
- text categorization
- learning algorithm
- text classification
- natural language processing
- scene classification
- semi supervised
- video sequences
- machine learning algorithms
- domain specific
- object class
- text mining
- object detection
- probabilistic model