IGM‐based perceptual multimodal medical image fusion using free energy motivated adaptive PCNN

Abstract

Multimodal medical image fusion merges two medical images to produce a visual enhanced fused image, to provide more accurate comprehensive pathological information to doctors for better diagnosis and treatment. In this article, we present a perceptual multimodal medical image fusion method with free energy (FE) motivated adaptive pulse coupled neural network (PCNN) by employing Internal Generative Mechanism (IGM). First, source images are divided into predicted layers and detail layers with Bayesian prediction model. Then to retain human visual system inspired features, FE is used to motivate the PCNN for processing detail layers, and large firing times are selected as coefficients. The predicted layers are fused with the averaging strategy as activity level measurement. Finally, the fused image is reconstructed by merging coefficients in both fused layers. Experimental results visually and quantitatively show that the proposed fusion strategy is superior to the state‐of‐the‐art methods.

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

“…Conventionally, the values of the PCNN parameters are selected based on successive trials, but recently Xu et al [28] presented different methods related to the selection of the PCNN parameters. Moreover, the other PCNN-based approaches are presented using a adapting linking parameter based on local contrast, entropy (EN), directional gradient, saliency, fractional dimension, local visibility and intensity of pixels or coefficients [16,[28][29][30][31][32]. They exhibited improved fusion performance, however, successive trial-based parameter selection is involved in those approaches.…”

Noisy Remote Sensing Image Fusion Based on JSR

“…Compared with a single pixel value in an image, the human visual system is more sensitive to the edges, direction and texture information of the image [38]. Based on this, some definition evaluation metrics including spatial frequency (SF), energy of Laplacian of the image (EOL), SML and so on, are employed to evaluate the fusion results.…”

Optimized CT-MR neurological image fusion framework using biologically inspired spiking neural model in hybrid ℓ1 − ℓ0 layer decomposition domain

“…In the past years, the selection of all free parameters was done based on successive experiments on a regular interval (or hit and trial approach) [5,12,27]. To overcome the limitation of manual selection, some fusion methods are developed based on the adaptive selection of few free parameters using local contrast, entropy, directional gradient, saliency, local visibility and intensity of pixels or coefficients [23,25,28,29,30,31,32]. They showed improved fusion performance, however, successive trials based parameter selection is involved in those approaches.…”