Exploiting The Textural Information Of Uav Multispectral Imagery To Monitor Nitrogen Status In Rice

This study investigates the potential of texture analysis from the UAV-based multispectral imagery in qualifying N status in rice crops. Texture features and textural indices extracted from the UAV images were correlated with N-related variables. Results indicated that textural indices performed significantly better than individual texture features in N-related variables estimation. Textural indices performed better in plant level N-related variables than that in leaf level. With regards to texture analysis, direction vertical to the row orientation was recommended in row-planted crops. Those findings confirm that texture analysis from UAV imagery is a reliable technique in crop N status monitoring.