Shear-Resize Factorizations for Fast Multi-modal Volume Registration (MP-P7)

Author(s) : Ying Chen (Queen Mary, University of London, UK) Pengwei Hao (Queen Mary, University of London, UK) Jian Yu (Beijing Jiaotong University, China)

Abstract : Intensity-based method works well for multi-modal image registration owing to its effectiveness and simplicity. However it needs a large amount of computation for geometric transformation. To accelerate the transformation, in this paper, we present two shear-resize matrix factorizations for general 3D linear transformation. A transform matrix can be factorized into three shears and a fixed resize, or four shears and a customizable resize. Shears can be implemented very fast by memory shift, and a resize can be done by axis-aligned resampling. The factorizations can be applied to both rigid-body and affine transformations. Their efficiency is performed by experiments on MRI and CT volumes. The methods are quite promising for hardware and parallel implementation.

Menu