Multi-Focus Image Fusion using Intersecting Cortical Model

Zhiping Xu and Jinhong Pan


ICM, Image fusion, Fire time series map


We present a novel method for multi-focus image fusion based on the fire time series map (FTM) of neuron excitation generated from the neuron outputs inside Intersecting Cortical Model (ICM). Compared with previous research achievements, there is no need to decompose input source images in our method. Our method utilizes the FTM difference between two images as the focus criteria to implement multi-focus image fusion. Images with different focus are directly fed into the ICM in order to calculate the FTM. Meanwhile focus measure is carried out for each source image based on FTM. According to the results of the focus measure, the proposed method decides how to fusion different parts from different source images. Three experiment groups are designed to testify the performance of the proposed method. Several state-of-art methods are compared with our method in this paper. Based on the results of the experiments, our method exceeds other methods known presently in both subject and object evaluation criteria.

Important Links:

Go Back