Journal ID : AMA-18-01-2023-11958
[This article belongs to Volume - 54, Issue - 01]
Total View : 462

Title : Estiation of dry mass and nitrogen content for soybean using ground-based hyperspectral imagery

Abstract :

In this research, hyperspectral imagery was applied to analyze the dry mass and nitrogen content for black and soybeans depending on vegetation growth stages. There were significant differences in dry mass (R3 and R5 stages) and nitrogen content (R1 and R3 stages) between black and soybeans. Moreover, the reflectance of black bean was higher than that of soybean when the reflectance was compared to each other. In the result of each vegetation growth model, the precision and accuracy were variable not only depending on vegetation stages but also depending on the variety. The influence of vegetation stage and variety was compared with the precision and accuracy of models based on the several combinations of data. In the variety combination PLSR models, the precision and accuracy using black and soybeans was better than that of growth stage combination PLSR model. Although the accuracy was variable depending on growth stages, it was possible to explain more than 84% of dry mass and 82% of nitrogen content for black and soybeans using hyperspectral reflectance. This result might be helpful to reduce the production cost and increase the self-sufficiency of edible soybeans in order to improve the efficiency of soybean production.

Full article