An evaluation of image fusion applied to inland and coastal waters


Image fusion algorithms can be used to increase the spatial resolution of images. Therefore, evaluate the performance of fusion methods in particular environments such as inland and coastal waters is necessary to ensure a better estimation of optically active components in waters with great spatial variation. Hence, our objective is to assess the performance of HSI, Wisper, and LMVM fusion methods in different water bodies. Our results showed the gradient of watercolor was the principal reason for uncertainties and all analyzed bands resulted in similar performances. Furthermore, HSI and Wisper failed to preserve the spectral and spatial resolution, respectively; and LMVM had better results in preserving spectral resolution and enhancing spatial details. Our results indicate that LMVM method has more potential application for image fusion in inland and coastal waters imagery.