Estimating chlorophyll content of rice leaves on different layers based on scanning images

Real-time and accurate method of diagnosing and monitoring crop nutrition is the foundation of scientific fertilization management. In this study, color characteristic information was acquired from scanning images of rice leaves at the jointing by MATLAB, through analyzing the relationship between chlorophyll content which were abstracted from leaves on different sites and 27 color characteristics, the result showed that flag (first) leaf can predict nitrogen content optimally, and H value was used to establish the best estimation model of chlorophyll content. Finally, the regression models were verified by using other samples, and RMSE of Chl a, Chl b, Chl (a+b) was 0.98, 0.42 and 2.10 respectively, It is concluded that the scanner could be utilized to predict chlorophyll content of rice leaves.