Abstract
A MULTI-RESOLUTION TECHNIQUE FOR THE IMAGE FUSION OF DISSIMILAR MODALITIES OF MEDICAL IMAGES USING MODIFIED DFB
D. Sheefa Ruby Grace* and Dr. L.Mary Immaculate Sheela
ABSTRACT
This paper presents an innovative multi-scale technique for fusing two dissimilar modalities of medical images. Image fusion is the process of combining two images to form a single image. The new image will be more informative than these two input images. The two dissimilar modalities are CT and MRI medical images. CT image provides better information about denser tissue and it is not good for soft tissues. MRI image provides better information about the blood vessels, brain, heart, spinal cord and other internal organs and it is less accurate for bones. These two images are fused into a new image to improve the information content for diagnosis. Varieties of algorithms such as Discrete Wavelet Transform, Pyramid Wavelet Transform, Stationary Wavelet Transform, Curvelet Transform and Contourlet Transform are available to fuse two images. This paper proposes an Innovative Contourlet Transform. The fusion rule, Select-Maximum is used. The work is implemented using MATLAB. The results are compared with the existing techniques by using various performance measures such as Image Quality Index (IQI), Mutual Information (MI), Fusion Index (FI),Fusion Symmetry (FS), Fusion Factor (FF), Standard Deviation (SD), Entropy (EN), Root Mean Square Error(RMSE), Peak Signal to Noise Ratio (PSNR).
[Full Text Article] [Download Certificate]