Increasing the Quality of MRI and PET Images Fusion by Combination of Gabor Wavelet and Retina Model

The goal of image fusion is integration of two or more images into a single image for increasing the accuracy and quality of interpretation of images. The functional imaging system such as PET (positron emission tomography) produces images with high spectral content but low spatial information. Images that achieved by the structural imaging systems such as MRI (magnetic resonance imaging) have good spatial resolution and low spectrum. The proposed fusion method in this paper extracts spatial features using Gabor wavelet from structural image and injects them into functional image. This algorithm is a combined method, which exploits the power of retina model in preserving spectral information. The proposed method is compared to IHS, Brovey, PCA, discrete wavelet transform, NSCT and retina model fusion techniques. For quantity assessment, four famed and traditional evaluation methods are used. The quality and quantity results show the proposed algorithm significantly improves the fusion performance.