Multi Model Image Registration and Fusion using Fast Discrete Contourlet Transform

This paper proposes Transform Domain Fusion Rule (TDFR) via high pass modulation using Local Magnitude Ratio (LMR) in Fast Discrete Contour let Transform (FDCT) domain. Contour let transform method uses low resolution multispectral image and Cartosat-1 Panchromatic (PAN) of spatial resolution 2.5m is used as high resolution panchromatic image. This both images are up sampled in order to resize the image and registered using mutual information to parameterize and solve the correspondence problem in feature-based registration. This fusion rule generates as high resolution multispectral image at 2.5m spatial resolution with the help of Inverse Fast Discrete Contour let Transform increases the contrast of an image and thus gives sharper visual appearance. Different techniques are compared with this method such as Wavelet, Principal component analysis (PCA) and Curve let also tabulated. Keywords—Fast Discrete Contour let Transform (FDCT), Local Magnitude Ratio (LMR), Panchromatic (PAN), Principal Component Analysis (PCA),Transform Domain Fusion Rule (TDFR).