Multimodal Image Fusion in Sensor Networks using Independent Component Analysis

We present a novel image fusion algorithm based on ICA that has an improved performance over sensor networks. It employs segmentation to determine the most important regions in the input images and consequently fuses the ICA coefficients from the given regions. Sparse coding of the coefficients in ICA domain is used to minimize noise transferred from input images into the fused output. Experimental results confirm that the proposed method outperforms other state-of- the-art methods in the sensor network environment, characterized by JPEG 2000 compression and data packetization.