Analysis of the fractal characteristic of the hyperspectral curves of the rice under plumbum pollution stress

Based on previous spectral parameters to describe the “large scale change” of the spectral curve proposed by remote sensing scholars, the objective of this paper was to propose the method of describing subtle change in hyperspectral curve for rice under plumbum (Pb) pollution stress by using fractal dimension (D), which was calculated by wavelet transform. For pollution-free rice and rice under Pb pollution respectively, the D of hyperspectral curves (450 nm∼700 nm) and second derivative curves were calculated at the tillering stage and the heading stage. Results showed that the D of two curves of the rice under Pb pollution all decreased, compared with those of the pollution-free rice. Meanwhile, it comes to conclusion that it is the best stage to monitor the rice under Pb pollution at the tillering stage. Finally, Least Squares Regression is applied to establish the relationship between the D of second derivative curves and chlorophyll concentration of the rice under Pb pollution at the tillering stage, with high prediction accuracy.