Morphology-based spatial filtering for efficiency enhancement of remote sensing image fusion

Abstract

Better visual interpretation and high-level feature extraction are always desired for remotely sensed image processing applications. The fulfillment is most prominent when a Multi-Spectral (MS) image fused with a PANchromatic (PAN) image for the same geographic location produces another MS image with added spatial resolution. Development of such fusion algorithms based on approaches like Component Substitution (CS) and Multi-Resolution Analysis (MRA) is continuously researched over the last few decades. In this paper, the authors propose the morphological operator-based image fusion algorithm featuring nonlinear decomposition. Also, the use of morphological operator based spatial filtering is successfully demonstrated for efficiency enhancement of the proposed and a few of the sophisticated image fusion algorithms based on CS, MRA, and state-of-the-art deep learning approach. The potential of the presented work is proved through reduced- and full-resolution assessment procedures utilizing two data sets acquired by WorldView-3 and WorldView-2 satellite sensors.