Superpixel generation for SAR images based on DBSCAN clustering and probabilistic patch-based similarity.
Hao HuBin LiuWeiwei GuoZenghui ZhangWenxian YuPublished in: IGARSS (2017)
Keyphrases
- sar images
- clustering algorithm
- synthetic aperture radar
- clustering method
- k means
- density based clustering algorithm
- density based clustering
- data clustering
- similarity measure
- oil spill
- sar imagery
- spatial clustering
- instance level constraints
- bayesian networks
- cluster analysis
- automatic target recognition
- speckle noise
- categorical data
- target detection
- sar image segmentation
- edge detector
- probabilistic model
- optical images
- sar data
- arbitrary shape
- feature selection
- image patches
- image data
- low level
- high quality
- face recognition
- polarimetric sar