Bayesian dictionary learning for hyperspectral image super resolution in mixed Poisson–Gaussian noise
作者：
Highlights：
•
• We propose a beta process analysis-based sparse representation regularization term.
• We learn the dictionary based on the reduced low dimension subspace.
• The dictionary learning and image estimating are unified into an formulation.
• We update the dictionary self-adaptively by variational Bayesian method.
摘要
•We propose a beta process analysis-based sparse representation regularization term.•We learn the dictionary based on the reduced low dimension subspace.•The dictionary learning and image estimating are unified into an formulation.•We update the dictionary self-adaptively by variational Bayesian method.
论文关键词：Hyperspectral image,Multispectral image,Mixed Poisson–Gaussian noise,Bayesian dictionary learning,Alternating direction method of multipliers
论文评审过程：Received 23 January 2017, Revised 8 September 2017, Accepted 8 September 2017, Available online 19 September 2017, Version of Record 26 September 2017.
论文官网地址：https://doi.org/10.1016/j.image.2017.09.003
