Fusion Of Mr And Pet Brain Images Based On Lab And Log-Gabor Transforms

Abstract

In restorative imaging, positron emanation tomography (PET) demonstrates metabolic changes of a life form in pseudo shading and attractive reverberation (MR) imaging presents anatomical structures in dim level.The standard data fusion methods may not be satisfactory to merge a high-resolution MR image and a low-resolution PET image because they can distort the spectral characteristics of the multispectral data..In medical imaging the fusion of PET and MR medical images produces a composite image inspecting the anatomical structures along with metabolic changes.The fusion methods,like standard IHS,PCA, and wavelet-based methods have been used for merging purpose.Our goal is to preserve both spatial (anatomical structures in MR) and spectral fidelity (colors in PET image)as much as possible.For this purpose , a new fusion method is proposed based on LAB color model and Log-Gabor wavelet transform.In our fused method, both anatomical structures and color changes of PET and MR images are rendered with effectively reduced color distortion.The LAB color model can result in merged images with improved intensity with respect to those obtained by standard HIS,PCA  and standard wavelet-based fusion methods.