Title :
An EM- and wavelet-based multi-band image restoration approach
Author_Institution :
Shaanxi Key Lab. of Inf. Acquisition & Process., Northwestern Polytech. Univ., Xi´an, China
Abstract :
This work presents a restoration approach with two available observations of the same scene, each with specific spatial and spectral resolutions. The proposed restoration approach is based on a multi-band image restoration scheme employing Expectation-Maximization (EM) algorithm and implemented in wavelet domain. It can be easily applied in hyperspectral and multispectral image fusion and restoration to acquire image with both high spatial and spectral resolutions. Simulation experiment is employed to validate the proposed approach. A performance comparison is made with its counterparts in image domain.
Keywords :
expectation-maximisation algorithm; geophysical image processing; hyperspectral imaging; image fusion; image resolution; image restoration; EM-based multiband image restoration approach; expectation-maximization algorithm; hyperspectral image fusion; hyperspectral image restoration; multispectral image fusion; multispectral image restoration; spatial resolutions; spectral resolutions; wavelet-based multiband image restoration approach; Digital signal processing; Image restoration; Noise; Remote sensing; Spatial resolution; Wavelet domain; Expectation-Maximizition (EM) algorithm; hy-perspectral image; multipectral image; restoration;
Conference_Titel :
Digital Signal Processing (DSP), 2014 19th International Conference on
Conference_Location :
Hong Kong
DOI :
10.1109/ICDSP.2014.6900738