Fusion of data for analysis of images recorded by satellites: Proposal of metadata model

Abstract

Several technologies have emerged in recent decades that have provided considerable growth in the volume and complexity of digital image repositories. In view of this development, it is also necessary to create new information treatment strategies, especially those related to organization and recovery. Historically, the use of metadata is a technique used to identify image files. In this sense, data fusion can be used as a method to connect parameters in heterogeneous data sets for indexing in image repositories, in order to establish connections. In this sense, the objective of this work is to demonstrate the potential of data fusion for the organization of images recorded by different satellites, specifically, to develop a metadata model for the representation of information in data fusion by AI systems. This research is an exploratory bibliographic review used to demonstrate the potential of data fusion from images recorded by different satellites, coming from two NASA monitoring systems: National Snow & Ice Data Center and the GES DISC Data Archive, with further development proposal of a metadata model to assist data fusion, by observing common parameters and exploring the metadata structure. At the end of this work, it was possible to prove that the fusion of image data stands out as an efficient strategy for the representation and recovery of information and collaborates to expand the understanding of climate change on the planet.