Novel vegetation estimation index computation in arid environments

Abstract

This paper proposes a novel vegetation index for the monitoring and study of vegetation in arid environments. Our aim is to help maintain a record of the area of vegetation available in relation to the total area of the specified region and provide an understanding of vegetation growth. The study uses JAI AD-080GE multi-spectral 2-channel CCD camera for multispectral image sample collection and computer vision techniques for automatic vegetation detection and segmentation. Compared with traditional vegetation indices, this index is less computationally complex while still supplying a robust approximation of the vegetation in the environment with an 96% accuracy. This vegetation index has many promising uses for vegetation estimation and its simplicity allows for continuous monitoring of vegetation growth among other applications.