MULTI MODAL MEDICAL IMAGE FUSION USING WEIGHTED LEAST SQUARES FILTER

B.S. Saini, Vivek Venugopal

Abstract: A novel multi modal medical image fusion method based on weighted least squares filter is proposed. To perform the image fusion, a two-scale decomposition of the input images is performed. Then weighted least squares filter is used to calculate the weight maps for the base and detail layers and then a weighted average of the base and detail layer is performed to obtain the fused image. The performance of the proposed method was compared with several other image fusion methods using five quality metrics based on information present (QMI), structural information retained (QY and QC), features retained (QG and QP) and it was found that the proposed method produces a robust performance for fusion of multi-modal images

Keywords: Medical Image, Multimodality, Image Fusion, Weighted Least Squares Filter

DOI: https://doi.org/10.15623/ijret.2014.0306024