We propose an automatic algorithm to analyse the swelling kinetics of individual particles floating in a diluted suspension. Our approach is illustrated with a case study in which a diluted aqueous suspension of starch granules are submitted to various heating treatments. The evolution of modified waxy maize starch granules placed on a temperature-controlled stage was tracked using time-lapse light microscopy (one image per second). We relied on an automatic algorithm, developed for the present study, to estimate morphological and kinetic parameters for a large number of granules. Observations showed that above a given temperature, starch granules will evolve towards an equilibrium size. Due to variability of swelling kinetic, individual behaviour was different from the global one. These results suggest that modelling works should take into account the size heterogeneity and kinetic variability of starch granules.