K-textures, a self supervised hard clustering deep learning algorithm for satellite image segmentation.
Fabien H. WagnerRicardo DalagnolAlber H. SánchezMayumi C. M. HiryeSamuel FavrichonJake H. LeeSteffen MauceriYan YangSassan SaatchiPublished in: CoRR (2022)
Keyphrases
- image segmentation
- learning algorithm
- spectral clustering
- clustering algorithm
- unsupervised clustering
- markov random field model
- clustering method
- k means
- unsupervised learning
- data clustering
- training data
- machine learning
- graph cuts
- fuzzy c means clustering
- graph partitioning
- cluster analysis
- data points
- supervised learning
- hierarchical clustering
- image processing
- multiscale
- texture images
- reinforcement learning
- image analysis
- segmentation algorithm
- remote sensing
- level set method
- image segmentations
- image segmentation algorithm
- deep learning
- computer vision
- texture segmentation
- semi supervised
- markov random field
- shape prior
- level set
- active contours
- learning problems
- machine learning algorithms
- deformable models