Angular Effect In Proximal Sensing Of Leaf-Level Chlorophyll Content Using Low-Cost Diy Visible/Near-Infrared Camera

Proximal sensing is increasingly highlighted, owing to its potential for various applications in precision agriculture. However, the uses of its related special devices like visible/near-infrared (VNIR) cameras for crop mapping are yet limited in practice, due to their high costs in market. To handle this issue, this case study was dedicated to modifying a low-cost RGB camera as a VNIR one in a do-it-yourself (DIY) way - replacing its infrared-blocking optical filter with a red-blocking one - for estimating leaf-level chlorophyll content. To enhance its performance, the angular effect that proved to impact the quality of imaging in conventional remote sensing was examined here. Specifically, in the cases of 120 individual leaves for four plant species, 24 vegetation indices (VIs) were proposed and derived from the three bands (blue, green, and near-infrared) of the collected images; then, the same operation was repeated for a set of leaf inclinations (theta, the angle between the optical axis of the camera and the normal vector of the leaf-fixed plate, set from 0 degrees, 5 degrees, ..., to 50 degrees). The results showed that for each of the four plant species, different optimal VIs for leaf chlorophyll content retrieval were detected for different geometries of leaf reflectance. This suggested that more modular retrieval models capable of integrally reflecting the angular effect best should be developed for accurately sensing of leaf biochemical content. Overall, the contribution of this study is of fundamental implications for advancing quantitative proximal sensing in plant biochemistry and precision agriculture.