Iron ore pellets are one of the major iron-bearing raw materials, along with lump ore and sinter, for the steelmaking processes. Pellets are produced from ultrafine fractions of iron ores, which were previously considered as tailings of mineral beneficiation. The porosity generated during the pelletizing process is an important characteristic of the material because it allows internal gas flow, increasing its reducibility and consequently the process efficiency. On the other hand, the porosity affects the physical strength of the pellets, which must withstand all loads during handling operations, transportation and metallurgical processes. Thus, the amount, size, shape and spatial distribution of pores are important features for the pellet quality control. Traditionally, most analytical techniques used to characterize the porosity of pellets are mercury intrusion porosimetry (MIP) and optical microscopy (OM). Nevertheless, MIP allows evaluating only pores connected to the surface, in addition mercury is volatile and toxic, offering risks to the environment and human health. OM, in turn, is limited to two-dimensional space and can reveal unrepresentative information. Both techniques are destructive and consequently prevent further analysis of the same sample. The present work proposes the development of a methodology for the tridimensional characterization of the porosity in iron ore pellets through X-ray microtomography (MicroCT) and image analysis in order to separately determine the different types of pores (open and closed). 25 samples provided by the Vale mining company were first analyzed by MicroCT and then by MIP or OM. For optimization purposes, some operating parameters of MicroCT were tested, such as the use of lenses, different geometric configurations, and the number of projections, which directly affect the obtained image resolution and the analysis time. Comparing the results obtained in MicroCT with the results obtained by MIP and OM in equivalent samples, smaller porosity measurements were observed for MicroCT, due to the poorer resolution of the system. However, this methodology has been able to separately quantify the open and closed porosity, to describe the spatial distribution of pores, and to measure their size and shape.