Seismic Fault Segmentation via 3D-CNN Training by a Few 2D Slices Labels.
Yimin DouKewen LiJianbing ZhuXiao LiYingjie XiPublished in: CoRR (2021)
Keyphrases
- training set
- image labeling
- image segmentation
- training examples
- cardiac mri
- fault diagnosis
- region growing
- segmentation algorithm
- semantic segmentation
- segmented images
- medical images
- level set
- multiscale
- fully automatic
- labelled data
- quantitative measurements
- segmentation method
- mouse brain
- image analysis
- training samples
- supervised learning
- shape prior
- fault detection
- segmentation accuracy
- cellular neural networks
- training data
- labeling problems
- seismic data
- ground truth labels
- structured prediction
- energy function
- training process
- object segmentation
- image regions
- labeling process
- class labels
- label field
- edge detection
- image registration
- high resolution
- ground glass opacity