In cardiac myocytes Ca2+ release events based on the intracellular second messenger inositol-1,4,5-trisphosphate receptor (InsP3R) activation are not conclusively characterized. This Ca2+ release mechanism may play a significant role under pathophysiological conditions, e.g. atrial arrhythmogenicity. In this study we characterize InsP3-induced SR-Ca2+ release events (Ca2+ puffs) and spontaneous SR-Ca2+ release events (Ca2+ sparks) based on RyRs openings in atrial myocytes. Local Ca2+ release events were examined in intact or permeabilized atrial myocytes acutely isolated from transgenic mouse hearts overexpressing InsP3Rs by using rapid 2-dimensional confocal imaging (150 fps). InsP3Rs were activated by rapid superfusion with InsP3 (permeabilized cells) or by ET-1 (intact cells). Xestospongin C was used as antagonist. Ca2+ puffs were solely appearing in the presence of intracellular InsP3. However, Ca2+ puffs may trigger Ca2+ sparks and both likely coexist, making the event separation and analysis from confocal images challenging. We compare two analytical approaches for quantitative Ca2+ event analysis based on different mathematical formalisms. One procedure operates with a spark as the rejection of the null hypothesis that the fluorescence change is noise. The second performs pixel-by-pixel fitting of the fluorescence signal and reconstructs underlying Ca2+ events via a clustering algorithm. Although rapid confocal imaging of the entire cell cross section always suffers from low signal-to noise ratio, this approach gives a more detailed and representative view of local Ca2+ events. Our new formalism for Ca2+ event analysis allows us to separate and characterize micro Ca2+ events (Ca2+ puffs, Ca2+ sparks) by their absolute number, amplitude, signal mass, area, FWHMX, FWHMY and kinetics using high speed imaging. We found that Ca2+ puffs exist as specific events in atrial myocytes, which could not be seen previously. Supported by SNF, SciEx and Novartis Res. Foundation.