Computation of signal-to-noise ratio of airborne hyperspectral imaging spectrometer

Remote sensing imaging spectrometry is widely used in fields such as atmosphere survey, mineral discovery, forest surveillance and so on. The signal-to-noise ratio (SNR) of an imaging spectrometer reflects its radiance response and relates closely to its acquired image quality. Due to limited throughput, current slit-based hyperspectral imaging spectrometer stands a dilemma between its spectral resolution and SNR. For an airborne hyperspectral imaging spectrometer, the SNR is required to be in the order of hundred while the intentional spectral resolution requirement is met. The evaluation of SNR is of significance to its overall design and indice determination. Varying from panchromatic camera, for an imaging spectrometer, not a single SNR value but many a value due to different wavelengths has to be worked out. The atmospheric transmission model is built and the shooting circumstance is simulated by use of the MODerate resolution TRANsmission (MODTRAN). The SNR reduction due to the input throughput cut caused by the entrance slit is specially taken into account and environmental factors such as flight height, solar altitude angle and surface albedo can be set as required. For an imaging spectrometer that works in both visible and near-infrared band, two detectors are adopted. Each one collects visible or near-infrared images separately and its noise is calculated individually. The resulting SNR of the imaging spectrometer of the whole working band can be used as the basis to carry out feasible airborne spectrometer indice determination.