FPGA-based smart camera for accurate chlorophyll estimations
Résumé
In this work, a new chlorophyll estimation approach based on the reflectance/trans-mittance from the leaf being analyzed is proposed. First, top/underside images from the leaf under analysis are captured, then, the base parameters (reflectance/trans-mittance) are extracted. Finally, a double-variable linear regression model estimates the chlorophyll content. In order to estimate the base parameters, a novel optical arrangement is presented. On the other hand, in order to provide a portable device, suitable for chlorophyll estimation under large scale food crops, we have implemented our optical arrangement and our algorithmic formulation inside an FPGA-based smart camera fabric. Experimental results demonstrated that the proposed approach outperforms (in terms of accuracy and processing speed) most previous vision-based approaches, reaching more than 97% accuracy and delivering fast chlorophyll estimations (near 5ms per estimation).
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...