A Survey on the Datasets and Algorithms for Satellite Data Applications

This survey compiles insights and describes datasets and algorithms for applications based on remote sensing. The goal of this review is twofold: datasets review for particular groups of tasks and high-level steps of data flow between satellite instruments and end applications from an implementation and development perspective. The article outlines the generalized data processing pipelines, taking into account the variations in data sources and intended use cases. It also provides an overview of the current satellite types, operational constellations, and the capabilities for onboard and ground processing. The review further compares popular datasets based on the specific objectives of their corresponding end applications. The comparison includes AI readiness information for the datasets. Particularly, between others, specification if they contain reproducible data splits or author's defined metrics. A study and explanation of the workflow are performed for the typical and experimental preprocessing pipelines and decision algorithms. These decision-making algorithms include artificial intelligence methods emphasizing deep learning algorithms for computer vision. A basic usage comparison of algorithms is performed for each defined task. In summary, the article presents the data flow from cameras and radars on satellite to end applications. It provides an in-depth analysis of selected scenarios that exemplify diverse approaches to extracting valuable information from data. These representative scenarios were picked to cover typical computational pipelines, for example, object detection or segmentation, and to list distinct approaches for obtaining versatile data-derived information.