Although filamentous microorganisms are widely used in industrial fermentation processes, their growth and differentiation are not yet fully understood, because their biomass is structured, and therefore difficult to describe and to quantify. This lack of appropriate tools can hinder the optimization and control of the fermentation. A quantitative image analysis method was therefore developed for characterizing the physiology of the penicillin-producing mold Penicillium chrysogenum. This method is based on a differential staining procedure showing six physiological states: growing material, three differentiated states characterized by an increasing granulation, a highly vacuolized state, and dead segments having lost their cytoplasm. The image analysis software, with versions written for monochrome and color images, consisted of a semiautomatic binary mask computation step and a fully automatic segmentation step based on a fuzzy classification.