A new class of prior models is proposed for Bayesian image analysis. This class of priors provides an inherent geometrical flexibility, which is achieved through a transformation of the coordinate system of the prior distribution or model into that of the object under analysis. Thus prior morphological information about the object being reconstructed may be adapted to various degrees to match the available measurements. An example of tomographic reconstruction illustrates the potential of this approach.
Keywords: maximum a posteriori (MAP) reconstruction, Bayesian estimation, warpable prior
Get full paper (pdf, 2098 KB)
Return to publication list
Send e-mail to author at kmh@hansonhub.com