基于改进Hausdorff测度的医学图像配准算法

A Medical Image Registration Algorithm Based on Revised Hausdorff Distance

本文针对在医学图像配准中采用Hausdorff(HD)测度算法时存在精度和计算量之间的矛盾,提出了HD测度的改进算法,在HD测度算法中首次引入非线性的Bayesian一致性函数,以达到提高配准精度和抗噪声性能的目的.实验表明,该方法有效地解决了采用传统HD距离测度算法时,因出格点和噪声点引起的误配问题,获得了较好的配准效果.

Aim at the inconsistency between precision and computing complication because of using traditional Hausdorff Distance （HD） in medical image registration, we proposed a new Revised liD measure which introduces non-linear Bayesian consistency criterion function for the first time to get the higher registration precision and the strong ability to reject noise. The experimental results demonstrate that the proposed algorithm overcomes the registration faults which were brought by the isolated nodes and the noise using traditional HD and good registration performances are obtained.