Skewed distribution of leaf color RGB model and application of skewed parameters in leaf color description model

Table 4 Constructed correlation models between the SPAD value and the leaf color parameters

Model	Fit type
F₁₌59.733 − 0.304 × R_Mean	Linear regression
F₂ = 76.134 − 0.441 × R_Mean − 11.203 × Y_Skewness − 1.516 × G_Kurtosis	Linear regression
F₃ = 19.38 + 7.972 × cos (1.314 × R_Median) − 6.747 × sin (1.314 × R_Median)	Fourier fitting
F₄ = 0.3344 + 0.8709 × R_Mean − 1 77.3 × R_Skewness − 0.005536 × R ²_Mean +2.8 76 × R_Mean × R_Skewness + 8.515 × R ²_Skewness − 0.0122 7 × R ²_Mean × R_Skewness − 0.1398 × R_Mean × R ²_Skewness + 7.301 × R ³_Skewness	Polynomial fitting

F₁: Using the mean parameters R_Mean, G_Mean, B_Mean and their combinations with a normality assumption to establish multivariate linear regression models by stepwise regression, then choosing the best model. F₂: Using all 20 parameters to establish multivariate linear regression models by stepwise regression, then choosing the best model. F₃: Using the Fourier function to fit and obtain the model. F₄: Using the MATLAB Curve Fitting Toolbox to fit the polynomial F4 that incorporates spatial bidirectional patterns

ISSN: 1746-4811