Algorithms
Obtaining ground truth is an expensive, laborious, and time-consuming task. For physical and economic reasons, few remote sensors measure actual conditions simultaneously. To fill this computational void, algorithms have been developed to extract atmospheric data directly from the spectra of individual pixels in the hyperspectral image.
An algorithm is a "step-by-step procedure for solving a problem in a finite number of steps that often involves repetition of an operation" (Online dictionary). In regards to remote sensing, algorithms are used to classify the pixels in an image.
To illustrate this very simply: The analyst knows for instance from in situ measurements, that the temperature in a certain place at the time when the satellite flew over, was 3 °C. The relevant pixel can then be classified as having 3 °C. The algorithm will then classify all other pixels in the image that contain the same information as this one, as having 3 °C. In another specified area, the temperature was measured as 5 °C, so the pixel in question is classified as 5 °C and so on, until the analyst has all the information he or she needs.