Abstract
Image fusion is the process of merging details from two or more images of the same scene taken at the same time or at different times to create more accurate images than the individual images might provide on their own. Pixel-based methods, decision-based methods, and feature-based methods are used in image fusion. In this work, shearlet transform based image fusion technique is proposed and developed. This proposed work is compared with wavelet transform based image fusion technique using SSIM and UIQI Values and the sample size for each group is 30. For sample size calculation threshold value is set to 0.05, confidence interval is 95%, power value is 80% and enrollment ratio is 1. The performance of image fusion is measured using two parameters namely Structural Similarity Index Measures (SSIM) and Universal Image Quality Index Measure (UIQI). High values of SSIM and UIQI indicates the better performance of image fusion algorithms. Shearlet transform provides mean SSIM values of 39.6000(dB),mean UIQI of 72.2567(%) and wavelet transform method provides mean SSIM values of 1.5090(dB),mean UIQI of 11.5283(%). Based on the experiments results from MATLAB software and from statistical analysis using IBM-SPSS software it is concluded that the shearlet transform based image fusion technique provides better results than wavelet transform based image fusion technique.

This publication has 8 references indexed in Scilit: