Real-time nitrogen detection system of corn crop using a multi-spectral imaging sensor

Nitrogen (N) is an essential nutrient required for plant growth and a major component of the chlorophyll molecule that enhances photosynthesis. However, excessive N fertilizer leaches into the groundwater and creates serious environmental problems. Thus, there is an opportunity for sensors that can assess plant N-deficiency throughout the growing season to enable producers to reach their production goals, while maintaining environmental quality through reduced fertilization.; A multi-spectral imaging sensor (MSIS) was developed for assessment of N-stress levels of corn (Zea mays L.) crops. The MSIS was a ground-based remote sensor for measuring leaf reflectance from a nadir view over a crop under natural ambient illumination. Images of three spectral ranges (green with 550 ± 50 nm, red with 650 ± 50nm, and near-infrared with 800 ± 50 nm) were used to derive a reflectance index based on narrow-band filter assumptions.; The sensing system was calibrated to determine the relationship between the MSIS reflectance response and a known target by setting the MSIS sensor, a sensor for ambient illumination, and the target reflectance panel normal to the solar position. Image intensity was maintained for all image channels by adjusting the camera exposure and gain using a fuzzy logic controller. The effect of varying solar zenith angles on stationary MSIS and ambient illumination sensor was successfully compensated and daily variation of the MSIS reflectance with response to a known reflectance panel was maintained for all three wavelength channels within a standard deviation of 0.83% reflectance in sunny conditions and 0.62% reflectance in cloudy conditions.; Real-time N-deficiency was estimated by using NDVI and mapped relative to unstressed healthy plants. The MSIS system provided a measure of crop response that correlated with plant SPAD measurements (−0.93 for G channel and −0.88 for R channel). The system had smaller measurement variation than the SPAD meter with a standard deviation of 0.13 in MSIS measurements whereas 0.26 in SPAD measurements. The MSIS-based supplemental N treatment improved the crop N status and increased the yield in one case, while in other case no yield improvement was observed with MSIS-based fertilization.