The determination of nitrogen value at various reading points on rice leaf using RGB imaging
Nitrogen is one of the important nutrients in rice cultivation. However, excessive application of nitrogen will attract pesticide, increase operational cost and may cause environmental pollution. Leaf colour is one of the indicators of nitrogen level. Hence, the applicability of RGB imaging system to determine the nitrogen status on rice leaf at different nitrogen levels and growth stages was investigated in this study. Four nitrogen treatments (0, 85, 170 and 250 kg ha-1) were arranged in a randomized complete block design (RCBD). Rice leaf images were acquired using an RGB imaging system and measurements using SPAD meter and colorimeter were collected as reference data. The acquired images were processed to extract the colour and pixel value using an image processing technique. Results indicated that there were significant differences in colour indices i.e., red (R), normalized red (r), intensity (V) and mean intensity (p=<0.001) at different nitrogen levels while blue (B), red (R), normalized red (r), normalized blue (b), Ikaw, average Hue (H), average saturation (S), average value (V) and maximum intensity (p<0.001) were significantly different for different reading points. The coefficient of determination (R2) values of the prediction models, which ranged from 0.527 to 0.998 indicate that RGB imaging can be used to estimate nitrogen at different levels and reading points and to detect colour changes in rice leaf.
Bachik, N.A., Hashim, N., Wayayok, A., Che Man, H. and Saipin, S. 2017. The determination of nitrogen value at various reading points on rice leaf using RGB imaging. Acta Hort. (ISHS) 1152:381-386
image processing technique, colour indices, spad meter, colorimeter, colour values