Deoxidation is an unavoidable step in the elaboration of steel. The study of its influence could improve the quality of low-carbon steel (0.20-0.25 wt.% of carbon). There are many deoxidation methods, and the most-common one consists of adding aluminum. Although it is a classic method, determining the optimal process parameters (quantity, yield, etc.) could be very sensitive. Deoxidation plays a determining role on inclusion cleanliness, especially on sulfide morphology. In order to control the efficiency of deoxidation, different techniques can be used. In this paper, an automated counting procedure on a scanning electron microscope with a field emission gun (FEG-SEM) is presented. This method was applied on samples cast in our laboratory under different deoxidation conditions. According to this, the resulting inclusion population is correlated with the aluminum content to find the optimal process parameters.