Advances in spectroscopy sensors have allowed the acquisition of ever-increasing volumes of data from scenes, either remotely, by air- or space-borne devices, or locally, by hand-held spectrometers or stand-alone cameras. With this boom in the amount of data available has also come a greater need for extracting useful information efficiently and for developing automated methods for novel applications. Traditional approaches to spectral analysis often require a great deal of human effort and prior knowledge, and have difficulty in processing high dimensional data sets provided by new sensors. This book, therefore, provides an alternative approach to select relevant features from hyperspectral data utilizing machine learning to automate the analysis. The methods are developed in the context of two applications: in biomedical imaging and in precision agriculture. The techniques discussed should be useful to graduate students and researchers in computer science and engineering interested in hyperspectral imaging, remote sensing or optimization for high dimensional data.