Application of Hyperspectral LiDAR on 3-D Chlorophyll-Nitrogen Mapping of Rohdea Japonica in Laboratory

Biochemicals, such as chlorophyll (Chl) and nitrogen (N), are closely related to photosynthesis process of vegetation. Their accurate estimation is an important topic in remote sensing Previous studies mainly focused on Chl-N content inversion leaf canopy level, few cared about their 3D distributions, which was also indicator for the growth status vegetation (GSV). Hyperspectral LiDAR (HSL) a novel active technology, has target-sensitive band with hyper-spectra resolution. Its point cloud data simultaneously contains rich spectral precise geometrical characteristics target. This work aims apply HSL mapping through constructing HSL-based indices (SIs). Except following SI forms previous works, normalized differential index (NDVI) ratio (RI) four-broad-band space were successively proposed invert whole based artificial neural network (ANN) method. These four broad-bands transformed relative response curve detector feature weights (FWs) multiwavelength respectively. Results show that most ANN models can accurately mean R2 >0.75, some fusing broad-band convolution transformation, namely FW-based RI, even obtain model 0.84 N inversion. Thus, be efficiently applied great potential GSV monitoring.