Evaluation of different vegetation discriminator indices and image processing algorithms to estimate water productivity

A canopy cover percentage is considered one of the most important evaluation criteria for evaluating the ‎simultaneous effects of impressive factors on water efficiency. Since digital cameras are developed and ‎widely available, the use of discrimination indices in the visible spectrum is making it possible to calculate ‎the leaf area index and chlorophyll content of vegetation covers. Therefore, in this study, the ‎performance of five plant Vegetation Discrimination Indices (VIDs) and a variety of thresholding ‎algorithms was compared in order to distinguish the sugar beet's vegetation cover from its background, ‎among which two new indices were introduced. In comparison with the old VID of Excess Green minus ‎excess Red (ExGR), using the new VID of Excess Green minus excess Blue (ExGB) and Riddler-Calvard's ‎thresholding algorithm resulted in a 29.54 percent increase in vegetation cover segmentation accuracy. ‎Following this step, we determined which function would best predictdry beet weight from vegetation ‎cover percentage, and the power function did the best. In order to estimate the yield, the segmentation ‎method based on Riddler-Calvard thresholding and the New Canopy Index of Vegetation Extraction ‎‎(CIVEn) had an error of 12.09 Kg. With an error of 41.25 Kg, the segmentation method based on Otsu ‎thresholding and ExGR index performed worst.‎