Comparative analysis of the hyperspectral vegetatation index and radar vegetation index: A novel fusion vegetation index

Abstract

The hyperspectral vegetation index (HVI) has shown promise in vegetation fields, but its relationship to the radar vegetation index (RVI) is not known in the context of various land covers. This work presents a comparative analysis of the HVI data derived from the AISA sensor and RVI values originating from the RADARSAT-2 quad-polarimetric synthetic aperture radar (SAR) data. Six types of land cover (buildings, forest, salt pond, tidal flat, ocean, and paddy field) were compared, and the patterns were investigated. In the RVI, forest areas show higher separability than the other types of land cover without exception. Also, in the HVIs, the forest areas indicate high values, without exception. The statistics of the region of interest (ROI) demonstrate that the RVI patterns of the six land-cover types are highly similar to those of the HVI. Thus, during bad weather conditions and at night, the RVI data could serve as an alternative to the HVI data. In addition to comparative analysis, we propose a novel fusion vegetation index (FVI) using the RVI and normalized difference vegetation index (NDVI). The proposed FVI creates obvious vegetation separation, more so than other land covers. Using the FVI, more effective vegetation monitoring could be possible in various vegetation monitoring fields.