Measuring Intra-Pixel Sensitivity Variations of a CMOS Image Sensor

Abstract

Some applications in scientific imaging, like space-based high-precision photometry, benefit from a detailed characterization of the sensitivity variation within a pixel. A detailed map of the intra-pixel sensitivity (IPS) allows to increase the photometric accuracy by correcting for the impact of the tiny sub-pixel movements of the image sensor during integration. This paper reports the measurement of the sub-pixel sensitivity variation and the extraction of the IPS map of a front-side illuminated CMOS image sensor with a pixel pitch of $6~\mu \text{m}$ . Our optical measurement setup focuses a collimated beam onto the imaging surface with a microscope objective. The spot was scanned in a raster over a single pixel to probe the pixel response at each (sub-pixel) scan position. We model the optical setup in ZEMAX to cross-validate the optical spot profile described by an Airy diffraction pattern. In this paper, we introduce a forward modeling technique to derive the variation of the IPS. We model the optical spot scanning system and discretize the CMOS pixel response. Fitting this model to the measured data allows us to quantify the spatial sensitivity variation within a single pixel. Finally, we compare our results to those obtained from the more commonly used Wiener deconvolution.