Directive Contrast based Multimodal Image Fusion in UWT and NSCT Domain

摘要

Objectives: The purpose of image fusion is to provide information integrated from different images to eliminate redundancy and contradiction between images. This paper presents medical image fusion using pixel level fusion with multi-level wavelet transform to obtain low frequency and high frequency subbands. This involves a pixel level averaging rule for appropriate fusion to integrate the decomposed image coefficients subbands. Methods: A two-stage multimodal fusion framework uses the cascaded combination of Un-decimated Wavelet Transform (UWT) and Non Sub-sampled Contour let Transform (NSCT) domains are used. Findings: This is to improve upon the shift variance, directionality, and phase information in the finally fused image. Applications/Improvements: A mathematical analysis of fused images is done using dedicated fusion metrics. The system performance is evaluated by using the parameters such as Peak signal to noise ratio, correlation and entropy.