中国虚拟人数据的配准算法

20世纪末出现的基于人体大量信息融合分析的精细化医疗技术,将完整人体的数字化信息和计算机技术相结合,使得基于实际人体信息的模拟和计算成为可能;到目前为止,中国虚拟人的研究刚刚开始。针对中国虚拟人数据的特点,提出了一种图像配准的新方法。根据图像的属性和用途,就配准的各个环节采用合理可行的方法,包括仿射变换、PV插值技术、均方差相似性测量、Powell优化、全局匹配方法。改进了算法流程框架,给出了由参考图像空间到浮动图像空间的逆向几何变换,避免了潜在的“空洞”问题。该算法流程框架将各个部分有机地结合到一起,高效、有序地对中国虚拟人数据图像进行配准。用该配准方法处理中国虚拟人数据,取得了令人满意的效果。

Registration Algorithm of Chinese Virtual Human Data

By the end of 20 century there appears the fine medical treatment technology based on analysis of mass information registration of human body.With the combination of digital human body data and computer science,this makes it possible that simulation and computation of actual human body information.By now there just begins to conduct research on Chinese Virtual Human.According to the characters of Chinese Virtual Human,we provide a new method for registration.Based on the attributes and functions of images,we adopt reasonable and applicable module for each part of this method including affine transform,partial volume distribution interpolation,mean square departure metric,Powell optimization and so on.We change the registration framework and convert forward geometric transform into backward geometric transform from reference space to floating space in order to avoid the 'hole' problem.We use registration framework to conglutinate all the parts together to register image efficiently and orderly.It is satisfactory to deal with Chinese Virtual Human data image with this method.