next up previous
Next: Learning the Joint Intensity Up: Introduction Previous: Maximization of Mutual Information

Incorporating a Prior Model

The framework for our registration process is illustrated in Figure 2. The method requires a pair of registered training images of the same modalities as those we wish to register in order to build the joint intensity model. To align a novel pair of images, we compute the likelihood of the two images given a certain pose based on our model by sampling the intensities at corresponding points. We improve the current hypothesized pose by ascending the log likelihood function.


  
Figure: Flowchart of the a priori training and the online registration.
\begin{figure}
\centerline{\psfig{figure=flowchart2.ps,width=3in}}\end{figure}



Michael E. Leventon
1998-09-30