Development of wavelet transform based image fusion technique with improved correlation on normal and thermal images for quality enhancement in comparison with discrete cosine transform based image fusion technique.

The area of image processing to which image fusion belongs. The goal of picture fusion is to create a single image that has more information than any of the individual photos used to create it, therefore improving upon the original images. The quality and usefulness of this information may be enhanced via the use of image fusion methods. From remote sensing to medical imaging, image fusion is an integral part of the image processing industry. In this study, a novel image fusion approach based on the wavelet transform was suggested and developed to enhance picture quality. Normalized correlation (NC) and universal image quality index (UIQI) values are used to evaluate both the proposed work and the Discrete cosine transform based image fusion approach, using 30 samples in each group. In the confidence interval has a 95% threshold, and the value is set at 0.05. Parameters normalised correlation (NC) university index quality measure presentation of picture fusion of each algorithm (UIQI). High values of NC and UIQI point to more successful image fusion. Mean values for NC and UIQI are 0.6359 and 1.0163%, respectively, for the discrete wavelet and cosine transforms (p=0.347 and p=0.639, respectively), whereas they are 0.6789 and 0.9496, respectively, for the discrete cosine transform. Discrete wavelet transform based picture fusion has been shown to outperform discrete cosine transform in experiments. [ABSTRACT FROM AUTHOR]