As described in Section 1.5.4, there are two operating modes for dispersionelement-based hyperspectral imagers: whisk-broom and push-broom. A pushbroom hyperspectral imager is a desirable form for Earth observations from space because it can achieve a higher SNR than its whisk-broom counterpart. This is because a push-broom hyperspectral imager does not need to scan ground samples individually in a cross-track line, and thus it has longer integration time to sense each ground sample and collect more photons. However, it carries the penalty of increased correction (another calibration task) difficulty. In a whisk-broom hyperspectral imager, all ground samples have their spectra recorded by a single linear detector array, one after another, whereas a push-broom hyperspectral imager, with a typical 1000 ground samples in a cross-track line, effectively has 1000 different linear arrays or spectrometers in need of calibration.
Online access to SPIE eBooks is limited to subscribing institutions.