A Duality Based Algorithm for TV-L1-Optical-Flow Image Registration
|Authors||Pock Thomas, Urschler Martin, Zach Christopher , Beichel Reinhard, Bischof Horst|
Proceedings of the 10th international Conference on Medical Image Computing and Computer Assisted Intervention
|Date||to appear 2007|
Nonlinear image registration is a challenging task in the field of medical image analysis. Intensity based registration uses the complete information found in the provided source images and generates a dense displacement field. In many applications discontinuities may be present in the displacement field, and intensity variations between two images may occur. Hence, our intensity based registration method employs a TV-L1 energy functional incorporating total variation regularization and a robust data fidelity force. In this work we propose a novel, fast and stable numerical scheme to find the minimizer of this energy. Our approach combines a fixed-point procedure derived from duality principles with a thresholding step. We show experimental results on synthetic and clinical CT lung data sets at different breathing states as well as registration results on inter-subject brain MRIs.