2-D sizing of sodium oxalate crystals by automated optical image analysis

A new automated static optical image analysis (SOIA) method was developed for characterising the shape of sodium oxalate crystals by separately measuring their length and width. The automated size analysis combines crystal image acquisitioning with a custom-developed numerical algorithm for morphological analysis of crystals. The automation of the two main measuring steps enables characterisation of a large number of crystals ensuring statistically meaningful results. As part of the numerical algorithm, a rectangularity-based morphological filter was implemented to eliminate overlapping crystals and crystal networks. The developed optical sizing technique has a low detection limit of 0.25 μm, which is close to the theoretical resolution of optical microscopy. It is demonstrated that the new automated SOIA sizing technique is very sensitive in identifying oxalate crystals of different shape. The measured crystal size distributions are compared to those obtained by an AccuSizer sizing instrument to confirm that they are consistent with the measurements obtained by a conventional crystal sizing instrument.