The main purpose of stereology is to extract quantitative informations from microscopic images. This make it possible to estimate geometrical parameters of three dimensional objects from lower dimensional samples. Statistical inference from a sample can be formulated in two different ways, depending on whether the sampling variation arises from intrinsic randomness in the population (model-based) or is assumed to arise from the random selection of the sample (design-based). These two approaches have very different practical implications, then there are two broad approaches to inference in stereology. In this thesis we discuss model-based and design-based stereology basic concepts and their applications, then we obtain the prarameters of stereology. Our next purpose is mathematically estimation of geometrical quantities such as volume and surface area for an object in three-dimensional space which is outstanding in statistics theory. Finally we estimate the parameters of pores in bread structure such that described as Boolean model.