Skip to main content

Table 7 Pre-processing methods

From: Evaluating the suitability of hyper- and multispectral imaging to detect foliar symptoms of the grapevine trunk disease Esca in vineyards

Method

Formula

Vector L2 normalization

\({R}_{\lambda }^{N}=\frac{{R}_{\lambda }}{\sqrt{\sum_{\lambda }{{R}_{\lambda }}^{2}}}\)

Vector SNV normalization

[56]

\({R}_{\lambda }^{N}=\frac{{R}_{\lambda }-\frac{1}{N}\sum_{\lambda }{R}_{\lambda }}{\sqrt{\frac{1}{N-1}\sum_{\lambda }{\left({R}_{\lambda }-\frac{1}{N}\sum_{\lambda }{R}_{\lambda }\right)}^{2}}}\)