Evaluation of Feed Value of IRG in Middle Region Using UAV
2017
Na, S.I., National Institute of Agricultural Sciences, RDA, Wanju, Republic of Korea | Kim, Y.J., National Institute of Crop Science, RDA, Milryang, Republic of Korea | Park, C.W., National Institute of Agricultural Sciences, RDA, Wanju, Republic of Korea | So, K.H., National Institute of Agricultural Sciences, RDA, Wanju, Republic of Korea | Park, J.M., National Institute of Agricultural Sciences, RDA, Wanju, Republic of Korea | Lee, K.D., National Institute of Agricultural Sciences, RDA, Wanju, Republic of Korea
Italian ryegrass (IRG) is one of the fastest growing grasses available to farmers. It offers rapid establishment and starts growing early in the following spring and has fast regrowth after defoliation. So, IRG can be utilized as the dominant/single species of grass used in a farming system, or to play a role as a large producing pasture and sacrificial paddock. The objective of this study was to develop the use of unmanned aerial vehicle (UAV) for the evaluation of feed value of IRG. For this study, UAV imagery was taken on the Nonsan regions two times during the IRG growing season. We analyzed the relationships between NDVIUAV and feed value parameters such as fresh matter yield, dry matter yield, acid detergent fiber (ADF), neutral detergent fiber (NDF), total digestible nutrient (TDN) and crude protein at the season of harvest. Correlation analysis between NDVIUAV and feed value parameters of IRG revealed that NDVIUAV correlated well with crude protein (r = 0.745), and fresh matter yield (r = 0.655). According to the relationship, the variation of NDVIUAV was significant to interpret feed value parameters of IRG. Eight different regression models such as Linear, Logarithmic, Inverse, Quadratic, Cubic, Power, S, and Exponential model were used to estimate IRG feed value parameters. The S and exponential model provided more accurate results to predict fresh matter yield and crude protein than other models based on coefficient of determination, p- and F-value. The spatial distribution map of feed values in IRG plot was in strong agreement with the field measurements in terms of geographical variation and relative numerical values when NDVIUAV was applied to regression equation. These lead to the result that the characteristics of variations in feed value of IRG according to NDVIUAV were well reflected in the model.
显示更多 [+] 显示较少 [-]