A non-destructive 3D MRI method combined with an image processing algorithm for characterization of individual eyes in semi-hard cheeses is proposed. The impact of MRI acquisition parameters allowing shortening of measurement time was analyzed to determine the optimal protocol providing accurate estimation of eye volume. The MRI imaging protocol was validated by comparison with ACT measurements. The method was applied to study eye development and growth in cheese during ripening. It was successfully used to evaluate number, volume and spatial distribution of eyes. Investigation of individual eye characteristics during ripening allowed differentiation of eye behavior. The MRI method made it possible to highlight the effects of the spatial heterogeneity of the cheese on eye growth rate. The principal application of this method would be for research and development purposes. (C) 2013 Elsevier Ltd. All rights reserved.