A Quantitative Analytical Framework for Photon Transfer Curve-Based Preflight Characterization of the Indian Remote Sensing Imaging Systems

Preflight performance characterization of spaceborne imaging systems offers high-quality images for scientific applications. A key challenge in the laboratory characterization process is to identify the candidate signal chains requiring performance optimization. For the Indian Remote Sensing (IRS) imaging systems, light transfer characterization has been identified as a standard process for performance characterization. Photon transfer curve (PTC) is another powerful and widely used tool to characterize the imaging systems in terms of camera gain (
$e^{-}$
/DN), read noise (
$e^{-}$
), charge-to-voltage conversion factor (CVF), signal-to-noise ratio, mean–variance linearity, and so on. There have been no detailed investigations on the PTC characteristics of the IRS imaging systems. We present here PTC-based characterization studies on two high-resolution IRS imaging systems, namely, Cartosat-1 and Cartosat-2. For this, a quantitative analytical framework has been developed, which enables comparative studies among multiple signal chains by applying various statistical measures on the PTC derived parameters. This framework provides not only a quantitative assessment of performance deviations but also enables performance traceability up to detector level. Taken together, our analysis shows that all the signal chains have well behaved PTC characteristics, and performance deviations are less than 10%. In particular, performance traceability is established by the close match of the system-level CVF values within the detector manufacturer’s specified range. Studies on the adequacy of linear approximation of the PTC curve reveal large residual errors in the lower dynamic range due to an increase in read noise floor. The analytical framework developed here can significantly help optimize future IRS imaging systems.