DeepCrackAT: An effective crack segmentation framework based on learning multi-scale crack features.
Qinghua LinWei LiXiangpan ZhengHaoyi FanZuoyong LiPublished in: Eng. Appl. Artif. Intell. (2023)
Keyphrases
- multiscale
- learning scheme
- learning systems
- learning algorithm
- learning tasks
- key features
- online learning
- scale space
- classification accuracy
- reinforcement learning
- low level
- supervised learning
- edge detection
- medical images
- prior knowledge
- learning process
- feature vectors
- test images
- image processing
- multiple features
- multiple scales
- feature relevance
- feature detection
- segmentation algorithm
- computer vision