Research Status and Prospect of Spatiotemporal Fusion of Multi-source Satellite Remote Sensing Imagery

Abstract

High spatial resolution monitoring of land surface and atmospheric environment dynamics requires high spatiotemporal resolution satellite remote sensing imagery as data support. However, the efficient, low-cost, and feasible solution is to blend the multi-source images with high-spatial and high-temporal resolution respectively to produce the desired high spatiotemporal resolution imagery required by different research or applications, which is subject to the limitations of satellite sensor' hardware technology and the budget constraints of launching more satellites. Although plenty of spatiotemporal image fusion research has been conducted, they are limited to specific data types, algorithm principles, application purposes, etc. Furthermore, the development of spatiotemporal image fusion algorithm presents a phenomenon of disorder. This study summarizes and generalizes the existing mainstream spatiotemporal fusion methods and classified them into four categories: ①spatiotemporal fusion based on land components; ②spatiotemporal fusion based on spatial information; ③spatiotemporal fusion based on temporal changes; (4)combined spatiotemporal fusion. Meanwhile, the study analyzes the problems and challenges faced by spatiotemporal fusion; and informs the prospects of the future development of spatiotemporal fusion method.