ABSTRACT
Modern agriculture faces a huge challenge in providing food security and safety to the ever growing world population while at the same time trying to protect the planet's environment and ecosystems. The development of precision agriculture represents the means to meet these challenges. Precision agriculture, or site-specific farming, aims at managing crops based on their specific needs. Remote sensing is an important part of a precision agriculture management system, with hyperspectral imaging as a powerful remote sensing tool thanks to its superior wavelength information, including both spectral coverage and bandwidth. An image's spatial, spectral, and temporal resolutions are critical to precision agriculture applications. Hyperspectral data have been collected from many platforms, including satellites, manned airplanes, ground vehicles, and unmanned aerial vehicles, for implementation in agriculture. To process hyperspectral imagery, numerous vegetation indices have been developed for different purposes. Multivariate statistical analysis and pattern recognition procedures are also widely used. This chapter presents an overview of hyperspectral imaging in agriculture with discussions on specific topics such as soil property and fertility sensing, herbicide drift detection, weed mapping, crop nitrogen stress detection, crop yield estimation, insect/pest infestation identification, and current trends in the use of unmanned aerial vehicles. It is expected that advances in hyperspectral sensor capability and computational power will continue to meet needs in current agricultural practices and that new applications will continue to be found in the future.
