Since post World War II and until 2008 the Danish pig producing industry (DPPI) has been in a continuing state of growth. In spite of an ever fiercer competition DPPI has managed to protect its position as export leader by maintaining a focus on research and development. Today, DPPI is in a state of recession and must increase the efficiency if not to reduce the production capacity further. The industry recognizes that a more efficient use of the raw materials is one of the largest and most important challenges. To meet this challenge it is a necessity to get a better understanding of the biological variation of pigs. The development of models for describing the biological variation of pigs is one of the key components needed to attain a better sorting of the pig carcasses and an improved cutting in the abattoirs. Such models can be related to possible products, which can be related to potential yield and order books. The Danish Meat Research Institute (DMRI) is currently constructing a representative database of virtual representations of pigs using X-ray Computed Tomography (CT). The database will serve as the foundation for the diversity modeling of pigs and for extracting predictors of quality and optimal use. This thesis integrates well-known techniques from the medical image analysis into the development of tools for automated analysis of the morphology of pigs. E.g. elastic image matching has been applied to establish spatial correspondence between the virtual representations of pigs in the database. The establishment of spatial correspondence is an essential preprocessing step for most automated analysis using the database.