Submitted to: Book Chapter
Publication Type: Book / Chapter
Publication Acceptance Date: November 18, 2010
Publication Date: N/A
Technical Abstract: Precision farming practices such as variable rate applications of fertilizer and agricultural chemicals require accurate field variability mapping. This chapter investigated the value of hyperspectral remote sensing in providing useful information for five applications of precision farming: (a) Soil management zoning, (b) weed control, (c) nitrogen stress detection, (d)crop yield estimation, (5) pest and disease control. When using remotely sensed hyperspectral data for soil management zone delineation, it was found that there were different sensitive regions in the electromagnetic spectrum (0.4 – 1.4 um) for different soil nutrient properties. For selective weed control, canopy reflectance in the spectral region from 450 nm to 900 nm, with emphasis on the region from red to NIR is important for weed and crop differentiation. Canopy reflectance has also been used in plant nitrogen stress sensing and yield estimation, as well as pest and disease detection. In plant nitrogen stress sensing applications, the most significant spectral region is the visible to NIR region. For crop yield estimation, it was generally regarded that canopy reflectance measured in the middle to late growing season gave the best prediction result. Lastly, this chapter summarized another potential precision farming application using hyperspectral data, which includes some research in detection and assessment of insect invasion, onset of disease, and fungal pathogen infection. In summary, hyperspectral remotely sensed data could be an important data source for field variability sensing and information extraction. This is a crucial step for the implementation of precision farming technology, which also consists of management decision making, precision field operation control, and result assessment.