Development of Wavelet Packet Transform based Image Fusion Technique with Improved PSNR on Medical Images for Simultaneous Evaluation in Comparison with Wavelet Transform based Image Fusion Technique

Abstract

Image fusion is the process of combining the characteristics of two or more images into a single image. Thisresearch project aims to develop transform domain-based image fusion techniques. In order to improve image quality, this work suggests and develops a wavelet packet transform-based picture fusion technique. Peak signal to noise ratio (PSNR) and universal image quality index (UIQI) values are utilised to compare the proposed work with a wavelet transform-based image fusion technique, with a sample size of 30 for each group. With a confidence interval of 95% and a threshold value of 0.05, the enrollment ratio is set at 1. To assess the effectiveness of image fusion, peak signal to noise ratio (PSNR) and universal picture quality index are utilised (UIQI). High PSNR and UIQI values imply enhanced performance of the image fusion process. In contrast to wavelet transform, which produces mean PSNR-l values of 11.5937 dB, 9.9557 dB, and 56.3563% UIQI, wavelet packet transform results in mean PSNR-l values of 17.1437 dB, 17.7810 dB, and 78.9833% UIQI. Based on the outcomes of the experiments carried out using MATLAB software and statistical analysis carried out using IBM-SPSS software, it can be concluded that the wavelet packet transform-based image fusion technique significantly out-performs the wavelet transform-based image fusion technique with <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$PSNR 1 \boldsymbol{(\mathrm{P}=0.003)},\ ,\ Psnr2\ \boldsymbol{(\mathrm{P}=0.001})$</tex> , and UIQI <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\boldsymbol{(\mathrm{P}=0.001)}$</tex> .