医学图像弹性配准中弹性模型的建立

Elastic model for medical image elastic registration

医学图像配准是医学图像处理中的一个重要研究课题,它是图像融合、图像与标准图谱的匹配、显微图像的重建等研究的基础。图像的配准方法有多种,它们可以分为刚性和弹性配准两大类。相对于刚性配准,弹性配准有着更高的精确性,而对于变形大的图像的配准,它是必须的。因此弹性配准的研究有着广泛的意义。本文根据图像的特性,结合弹性力学的理论和方法,建立了一种用于精确配准的弹性数学模型,并用这一模型进行图像的弹性配准,取得了较好的效果。

Image registration is an important subject in medical image processing. It is fundamental for image matching with atlas, image fusion and micrograph reconstruction. There are two categories of methods for image registration: rigid and elastic. Elastic methods can get more accurate result than rigid ones,especially for the images with large deformation. So it is necessary to conduct elastic registration under some cases.Based on the image's character and the elastic theory, an elastic model is proposed in this paper. This model is used for elastic image registration and satisfactory results are gained.