Orthogonal Subspace Projection-Based Go-Decomposition Approach to Finding Low-Rank and Sparsity Matrices for Hyperspectral Anomaly Detection.
Chein-I ChangHongju CaoShuhan ChenXiao-Di ShangChunyan YuMeiping SongPublished in: IEEE Trans. Geosci. Remote. Sens. (2021)
Keyphrases
- anomaly detection
- low rank
- hyperspectral
- eigendecomposition
- matrix completion
- regularized regression
- low rank matrix
- singular value decomposition
- rank minimization
- affinity matrix
- high dimensional data
- data matrix
- remote sensing
- singular values
- convex optimization
- multispectral
- infrared
- missing data
- tensor decomposition
- high dimensional
- linear combination
- matrix factorization
- semi supervised
- image data
- high order
- subspace clustering
- kernel matrix
- low dimensional
- dimensionality reduction
- sparse representation
- principal component analysis
- data sets
- unsupervised learning
- tensor factorization
- original data
- linear subspace
- feature selection
- learning algorithm
- machine learning