Interval type-2 fuzzy set based block-SBU for image fusion technique
J. Reegan Jebadass,P. Balasubramaniam
DOI: https://doi.org/10.1016/j.asoc.2023.110434
2023-05-22
Applied Soft Computing Journal
Abstract:Image fusion is the process of examining one or more obscure images, combining only the critical information in those images to make them a good quality image. Due to the large gap between the high dynamic range of landscape scenes and the low quality of consumer quality cameras, a single-shot image can't record the information of a single set. A part of image processing called image fusion is used to overcome the above problem. In this method, authors propose a new technology for image fusion with intuitionistic fuzzy sets. During the processing, the given images are converted into fuzzy images and then intuitionistic fuzzy images (IFIs) . Thus, a significant change occurs between the given image and IFIs . The resulting IFIs are converted into interval type-2 fuzzy images (IT2FIs) to overcome this problem. The proposed technique is compared to other procedures such as discrete cosine harmonic wavelet transform (DCHWT) , multi-resolution single value decay (MSVD) , primary component analysis (PCA) , standard wavelet transform (SWT) , two scale image fusion (TSIF) , and wavelet transform (WT) . The proposed method gives the best results based on performance analysis such as Entropy, standard deviation (STD) , average gradient (AG) , mutual information (MIF) , and blind referenceless image spatial quality evaluator (BRISQUE) . Using the proposed technique, one can obtain the best results for entropy, STD, AG, MIF, and BRISQUE respectively as 7.6241, 49.6745, 69.2100, 5.7276, and 3.3091. Hence it reveals that the proposed method performs better than other methods regarding overall visual quality and performance measurements.