Abstract
In this research, the spatial variability of amylose and protein content (hereinafter referred to as rice taste elements), using two test fields with different nitrogen fertilizer applications, was investigated. In test field 1, nitrogen fertilizer was applied with seven nitrogen treatments at both basal dressing and topdressing. In test field 2, nitrogen fertilizer was applied with seven nitrogen treatments at basal dressing and with variable rates at topdressing, depending on the plant growth. Although the rice taste and those elements showed a negative correlation, the applied amount of nitrogen fertilizer at topdressing showed a positive correlation with rice taste elements. Rice taste element prediction models were established using the rice taste elements in test field 1 at harvesting with reflectance of multi and hyperspectral remote sensing in the heading stage. The rice taste element prediction model was validated using the rice taste elements in test field 2. RMSEC (RMSEP) of models using multi and hyperspectral remote sensing were 0.774 (0.883) and 0.117 (0.122) at the model of amylose, 0.862 (0.965) and 0.136 (0.121) at the model of protein.
