Availability of Spectral Indices Based on RGB Image Data Obtained by Low-Cost UAV: Case Study

Abstract

Spectral indices, mostly based on multispectral image data, are vital tools in remote sensing applications to extract valuable information about the surface. One modern remote sensing tool is an unmanned aerial vehicle (UAV). UAVs can provide accurate image data on demand and carry multiple camera types. However, a limitation of low-cost UAVs is that they obtain only RGB image data. The RGB image data are based on the red, green, and blue parts of the visible light. Spectral indices can enhance spectral characteristics from RGB image data obtained by a UAV. This study used RGB image data of the Baroch nature reservation obtained by a low-cost UAV. Significant values of spectral indices were found in the RGB image data. The following spectral indices were used in this study: BCC, ExG(I), DSWI4, GCC, GLI, IKAW, MGRVI, MRBVI, NDYI, NGRDI, RCC, RGBVI, RGRI, RI, and VARI. Selected spectral indices based on calculation with RGB bands are used to process the collected imagery and identify the most common types of land cover: vegetation, bare land, and water.