From pyramids to state-of-the-art: a study and comprehensive comparison of visible-infrared image fusion techniques

Image fusion has emerged as a major area of research in the past few decades due to its extended applications. While progressing in the field of image fusion, a large number of techniques-based image transforms and spatial filters have been devised for both general and specific sets of images. The primary criterion of image fusion technique is to deliver high-quality visual perception besides giving a considerable objective evaluation rate. In this study, an information fusion rate-based study is done on recent, most researched, and high-performing state-of-the-art techniques using visible and infrared image datasets. These techniques have been chosen carefully, owing to their superiority in performance on both objective and subjective scales of evaluation and have been discussed in terms of their respective advantages and disadvantages. It is clearly evident that some rather primitive techniques can perform well as well as techniques based on a hybrid of various domains can significantly boost the information fusion rate.