Optical and SAR images Combined Mangrove Index based on multi-feature fusion

Abstract

The mangrove ecosystem is one of the most productive ecosystems, and it has unique ecological functions and great social and economic value. Accurate mangrove mapping is very important for mangrove dynamic monitoring and management. Many studies have yielded good results in mangrove mapping, and the optical remote sensing images are used as the main data source for mangrove index construction and mangrove mapping. However, less information is available for optical images affected by clouds and fog. This study constructed an Optical and SAR images Combined Mangrove Index (OSCMI) based on the idea of multi-feature fusion. An OSCMI-based classification scheme for rapid and accurate mangrove mapping was proposed, and the effect of tidal inundation is considered. We have carried out extraction experiments in four different types of typical mangrove areas in China using Sentinel-1 SAR and Sentinel-2 optical images. Several groups of comparative experiments were conducted to verify the superiority of OSCMI and the necessity of VV polarization mode. The results show that OSCMI performs well in the four typical mangrove areas of different types and has great application potential in mangrove large-scale mapping. And the introduction of VV polarization mode information is helpful to improve the accuracy of mangrove extraction.