About this document
Denoising Hyperspectral Images: Mixed Noise by Kranthirekha Chennaboina is a document available to read on EtoBox.
The document proposes a method for denoising hyperspectral images corrupted by mixed Gaussian and impulse noise. It formulates the problem as a synthesis prior optimization that exploits the spatial and spectral correlations in hyperspectral data. The optimization is solved using split-Bregman to estimate the clean hyperspectral image and impulse noise components. Experimental results on synthetic and real noisy data show the method achieves over 4.5 dB better peak signal-to-noise ratio and 40% better struc
- Author
- Kranthirekha Chennaboina
- Language
- EN