Effect of Farmland Consolidation on Protein Content of Paddy Rice

— Evaluation by UAV with multispectral camera —

Abstract

In this study, we developed a model for estimating the protein content of rice from UAV images of large size paddy fields. The correlation between the protein content and the vegetation indices was examined, and the GNDVI at 38 days after heading showed the highest correlation coefficient. A simple linear regression analysis using the GNDVI at 38 days after heading showed good estimation accuracy. The protein content estimation map created using the estimation model showed that the protein content differed depending on the plot before land readjustment and was higher in the banking part than in the cutting part. As well as the protein content, the amount of available soil nitrogen was higher in the banking part than in the cutting part. These results suggest that the protein content difference in these fields was caused by the spatial difference in the amount of available nitrogen due to land readjustment. In addition, it was suggested that the unevenness of the protein content due to the farmland consolidation influence could be observed by using UAV.