Journal of Pattern Recognition Research

In recent years the region based image fusion algorithms has drawn the attention of researches because it is more meaningful to combine regions rather than pixels. In this paper, a novel region based image fusion technique based on discrete wavelet transform using high boost filtering is proposed. The proposed algorithm allows us to achieve an accurate segmentation for region based fusion using graph based normalized cut algorithm. The resultant segmented image produced using this framework is used for extracting the regions from the input registered source images which are further processed to fuse differnt regions using different fusion rules. Proposed method is applied on the large number of registered images of various categories of multifocus and multimodality images and results are compared using standard reference based and nonreference based image fusion parameters. It has been observed from simulation results that the proposed algorithm is consistent and preserves more information compared to earlier reported pixel based and region based methods.