DYNAMIC CALIBRATION AND IMAGE SEGMENTATION METHODS FOR MULTISPECTRAL IMAGING CROP NITROGEN DEFICIENCY SENSORS

Abstract

Site-specific variable-rate nitrogen application is one of the core operations in precision crop management. The determination of an appropriate nitrogen application rate relies greatly on the capability of assessing crop nitrogen stress. A machinery-mounted multispectral imaging sensor has been developed for real-time crop nitrogen deficiency detection on the sprayer during fertilization operations. While field tests indicated that this image-based sensor was capable of detecting crop nitrogen deficiency “on the go,” the test results also showed that this sensor was very sensitive to ambient light changes and needed a considerably long image processing time to extract crop nitrogen deficiency data. To solve these problems, the research has developed a dynamic calibration method to compensate for ambient illumination variation on crop canopy reflectance, an image segmentation algorithm to eliminate the soil background noise, and a correlation model to estimate the SPAD values from the calibrated multispectral crop canopy reflectance. Field validation tests demonstrated that the developed sensor calibration and image processing algorithms improved the performance of the multispectral sensor on detecting corn nitrogen stress. Using the modified sensor resulted in a reasonable correlation between the estimated and measured SPAD values (R2 > 0.72). This research confirmed that it is technically feasible to design a machinery-mounted multispectral imaging sensor to detect crop nitrogen stress reliably and accurately.