基于演化建模的图像配准模型

Image registration model based on evolutionary modeling

传统的多项式配准模型原理直观,计算简单,但由于采用多项式建立模型,往往带有局限性,对于复杂的配准可能无效;传统的物理配准模型虽然配准精度较高,但对成像的各种因素需要精确掌握,实际操作难度较大。演化建模只需要根据问题的特征来确定模型结构的一些基本组成单元,就可以自动建立较为精确的模型。针对以上传统图像配准模型的不足,并考虑到演化建模具有智能挖掘模型的优点,提出了一种基于演化建模的图像配准模型。最后给出了该方法在图像配准中的实验,实验结果表明该方法有较高配准精度,适用于图像配准。

Traditional image registration model can be categorized to general model and physical model.General model is easy to compute,but usually inaccurate.The reason is limitation of using polynomia to build model.Physical model is accurate,but it calls for a lot of specialty knowledge.Evolutionary modeling can build accurate model automatically by evolutionary algorithms.It only needs to confirm a few elementary model structure units for the problem.Considering the shortcomings of traditional imaging registration model and the advance of evolutionary,this paper proposed a new image registration model based on evolutionary modeling.An image registration experiment is given in the last.Both of general model and evolutionary model are tested.The experiment result indicated that the evolutionary model has high precision for image registration and it is fit for image registration.