CrackGAN: A Labor-Light Crack Detection Approach Using Industrial Pavement Images Based on Generative Adversarial Learning.
Kaige ZhangYingtao ZhangHeng-Da ChengPublished in: CoRR (2019)
Keyphrases
- image analysis
- unsupervised learning
- input image
- learning process
- ground truth
- image classification
- edge detection
- object recognition
- learning algorithm
- detection method
- feature points
- supervised learning
- image features
- prior knowledge
- image retrieval
- image data
- information technology
- image database
- partial occlusion
- image segmentation
- learning tasks
- automatic detection
- object detectors
- false detections
- digital mammography
- image collections
- image annotation
- detection rate
- reinforcement learning
- three dimensional
- image processing
- e learning