Image fusion encloses all data analysis strategies aiming at combining the information of several images obtained with the same platform or by different spectroscopic platforms. Image fusion can imply building a single multiset or multiway structure with all images involved or connecting the related individual images through regression models. In any case, data analysis performed on structures of fused images always overcomes the results coming from individual image analysis. Image fusion may cover many diverse scenarios and purposes ranging from characterization of constituents in 3D hyperspectral images or in sets of related 2D images, process modeling, quantitative analysis, or simply taking advantage of the complementary spectroscopic information of different spectroscopic platforms. Trends on image fusion research involve developing strategies to overcome differences in spatial resolution among platforms keeping the relevant spectroscopic information and the best spatial sample description. With the aim of improving the instrumental spatial resolution of images, the image fusion idea is also present in superresolution algorithms that achieve its goal by combining information from images collected on the same sample and shifted by a subpixel motion step or from image frames collected in short time lapses in single molecule fluorescence imaging.