Image Fusion Algorithm for Medical images using DWT and SR

Abstract

Diagnosis of ailments require accurate information from same modality images or different modality images like Magnetic Resonance Imaging (MRI), Computed Tomography (CT), Positron Emission Tomography (PET) etc which can be obtained by Image fusion technique In Image fusion, there are various methods implemented based on Discrete wavelet transformation. The image after fusion will have significant and accurate information from different images than the individual images. The main advantage of image fusion is that, it increases the quality of the particular image and also reduces redundancy and randomness. In this paper, transform method and sparse representation method are implemented for Image fusion of MRI, PET and CT images of brain regions to obtain a better entropy. Discrete Wavelet Transform (DWT) method is applied to obtain low pass and high pass patches. The low pass patches undergo Sparse representation (SR) to form the fused patch of the image. The Max Absolute rule is applied to the high pass patch to form single coefficients patch. The patches from low pass and high pass fusion are combined using inverse DWT reconstruction to form single fused image. Various parametric values Elapsed time, Entropy, Standard Deviation and Mean are evaluated. Entropy is measurement of information content. Higher the entropy value means, more the detailed information in the image. The entropy of the proposed method is 2.9299 which is significantly higher than state of art.