A Remote Sensing Image Fusion Algorithm Based on Ordinal Fast Independent Component Analysis

Data fusion on remote sensing is hot in current image processing. A new approach, that is the ordinal fast independent component analysis for remote image fusion between Landsat ETM+ panchromatic and CBERS multi-spectral images, is proposed to eliminate high-order image data redundancy for two different remote sensing images. The independent components are done factor analysis, and then the fused image is obtained by applying image fusion rule. Visual and statistical analysis proves that the concept of fusion based on the ordinal fast independent component analysis is promising, and it significantly increases the signal-to-noise ratio and improves the fusion quality.