Laplacian Redecomposition for Multimodal Medical Image Fusion

NSST domain CT–MR neurological image fusion using optimised biologically inspired neural network

“…The primacy of the proposed method for CT–MR image fusion is highlighted by comparing its qualitative and quantitative performance with existing state‐of‐the‐art medical image fusion methods such as NSCT–PCNN by Das and Kundu [13], NSST–PCNN‐based medical image fusion proposed by Gupta [22], the sparse representation‐based method by Liu et al [33], parameter adaptive parameter adaptive pulse coupled neural network (PAPCNN)–NSST method by Yin et al [31], convolutional neural network‐based method proposed by Liu et al [34] and SD activated PCNN method in the NSCT domain presented by Li and Zhao [14]. Furthermore, to compare the fusion performance of the MR–SPECT neurological images, three recently developed image fusion methods such as the NSST domain bounded measure PCNN‐based method given by Tan et al [23], NSST‐domain weighted parameter adaptive dual‐channel PCNN‐based method by Panigrahy et al [32] and Laplacian re‐decomposition‐based fusion approach proposed by Li et al [7] are also considered.…”

“…13 and 14, it can also be visualised that the method proposed by Tan et al [23] does not preserve the structural details well and also results in over enhancement of the functional information which shows visual inconsistency in some regions of the fused images. The method proposed by Li et al [7] helps to preserve the anatomical information along with the limited reflection of colour information in the resultant images.…”

“… Visual comparison of fused images (pair‐1) obtained by (left to right: first row) Liu et al [ 34 ], Yin et al [ 31 ], Li and Zhao [ 14 ], Tan et al [ 23 ], (left to right: second row) Panigrahy et al [ 32 ], Li et al [ 7 ], Proposed method …”

“…Multi‐scale transform‐based fusion approaches are extensively used because of their ability to extract both the spatial and spectral information, which provides better localisation to spatial structures. Laplacian pyramid and its variants‐based fusion are one of the initial fusion approaches presented in the literature [6, 7]. These methods perform better than spatial domain fusion but lack directional selectivity causing serious blocking effects.…”

Edge Information Based Image Fusion Metrics Using Fractional Order Differentiation and Sigmoidal Functions

“…The robustness of the proposed metrics on the gray-scale and pseudo color medical image fusion is discussed using three recent image fusion methods, namely, information of interest in local Laplacian filtering domain (II-LLF) [62], phase congruency and local Laplacian energy in NSCT domain (PC-LLE-NSCT) [63], and Laplacian re-decomposition (LRD) [64]. In II-LLF, LLF is first applied to source images, then the fused approximate image is obtained using a maximum local energy (MLE) rule whereas the fused residual images are generated by an information of interest based scheme and MLE rule.…”