The implementation and performance of a connected regions centroiding algorithm for imaging photon counting

Many different types of imaging photon counting detectors use a centroiding algorithm to recover spatial resolution from a limited number of spatial sampling points. Typically, these use a fixed window about each photon event resulting in fixed patterning errors in the accumulated image due to neglecting the signal outside the window. We discuss a novel approach to the centroiding process using the connected region around each photon event. We describe the implementation and performance of this connected regions centroiding algorithm in an intensified CCD Photon Counting System (RALPCD).