Authors : D. Egfin Nirmala, R.K. Vignesh and V. Vaidehi
Abstract: In this study, a Pixel Image Fusion Method that combines multiple sensor images using Non-Subsampled Contourlet Transform (NSCT) and fuzzy Logic is proposed. The objective of this study is to improve the efficiency and performance NSCT based Fusion Method using fuzzy logic. As the Wavelet Transform Based Methods widely used for fusion can not represent smooth contours and edges in images accurately, the proposed method uses NSCT for multiresolution decomposition. The multisensor image information from different sensors is often complementary, uncertain, vague and redundant. As fuzzy logic is a powerful tool that can handle the uncertainty associated with vagueness, it is employed to determine the fusion weight for each NSCT low-pass coefficient based on the activity measures spatial frequency and entropy. In order to combine the NSCT high-frequency directional coefficients the local energy measure is employed. Experimental results demonstrate that the fused images of the proposed NSCT-Fuzzy scheme are with best visual quality and achieve higher scores than the Multiresolution-Based algorithms Discrete Wavelet Transform (DWT), NSCT and hybrid NSCT in terms of the fusion evaluation metrics.
D. Egfin Nirmala, R.K. Vignesh and V. Vaidehi, 2014. A Fuzzy Based Multiresolution Method for Multimodal Image Fusion. International Journal of Soft Computing, 9: 169-177.