Three automated techniques were developed for the alignment of Neuro-Images acquired during distinct scanning periods and their performance were characterized. The techniques are based on the assumption that the human brain is a rigid body and will assume different positions during different scanning periods. One technique uses three fiducial markers, while the other two uses eigenvectors of the inertia matrix of the Neuro-Image, to compute the three angles (pitch, yaw and roll) needed to register the test Neuro-Image to the reference Neuro-Image. A rigid body transformation is computed and applied to the test Neuro-Image such that it results aligned to the reference Neuro-Image. These techniques were tested by applying known rigid body transformations to given Neuro-Images. The transformations were retrieved automatically on the basis of unit vectors or eigenvectors. The results show that the precision of two techniques is dependent on the axial resolution of the Neuro-Images and for one of them also on the imaging modality, while the precision of one technique is also dependent on the interpolation. Such methods can be applied to any Neuro-Imaging modality and have been tested for both fMRI and MRI.