| Abstrakt: |
Unmanned aerial vehicle (UAV) equipped with RGB (colored image), multispectral and thermal camera, has excellent potential for making a revolution in the field of agriculture for determining water stress, nitrogen status, and fertilizer application for different crops in different climatic conditions. The main objective of this study is to develop relationships between vegetation indices obtained from RGB camera mounted on drone and ground truth data for assessment of nitrogen status in wheat crop. For this purpose, field experiments for wheat crop are conducted at Agricultural Research Farm, IIT Kharagpur, West Bengal, India. Two cultivars of wheat crop, namely Sonalika and PBWB-343, are used for the experiment. To determine the ground truth data, soil plant analysis development (SPAD) meter and Greenseeker are used to measure chlorophyll content and normalized difference vegetation index (NDVI), respectively. An RGB (red, green, blue) camera mounted on an UAV is flown over the Agricultural Research Farm at 20-m altitude to capture images of wheat field in Rabi season. The fuzzy c mean clustering method of image analysis is employed to determine 13 spectral indices (R, G, B, R/(R+G+B) (normalized R), G/(R+G+B) (normalized G), B/(R+G+B) (normalized B), normalized green-red difference index (NGRDI), ExG, ExR, ExGR,) from RGB image of wheat crop. This method consists of segmentation and successive analysis of the foreground color, i.e., only green plant parts. Relationships are developed using linear regression method between SPAD and NDVI, spectral indices and SPAD, and spectral indices and NDVI for wheat crop. Results show that NGRDI is highly correlated with NDVI and SPAD values. Coefficients of determination (R2) for NDVI and SPAD value, NGRDI and NDVI, and NGRDI and SPAD value are estimated as 0.55, 0.68, and 0.31 for PBWB-343 variety and 0.51, 0.49, and 0.43 for Sonalika variety of wheat. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink