摘要
基于标记的图像配准是现实中最常用的配准方法之一。在配准过程中,一般使用完全固定在头骨皮肤上的外部基准标记。基于模板的全自动图像配准的关键在于无需人工选择即可确定基准点的坐标。提出了一种基于模板的颅脑图像自动基准定位算法和技术,具有快捷、自动、准确的特点,实施过程包括以下三个步骤:边缘图构建,基于曲率的候选基准点检测,基于模板的基准点检测。实验结果说明该方法有助于实现基于基准点的全自动图像配准,相较于专家人工选择,具有广阔的应用前景。
Landmark-based image registration is one of the most popular registration methods in practice.During registration,we use external fiducial markers that are rigidly attached to the skin and to the skull.Determining the coordinates of the fiducials without human selection is crucial for a fully automatic template-based image registration.In this paper,a technique for automatic fiducial detection is presented.The technique,featured as fast,automatic and accurate,has three major steps:edge map construction,curvature-based fiducial candidate detection,and template-based fiducial detection.It will be useful for automatic point based registration.Experimental results indicate promising application compared to the manual selection by experts.
作者
王德兵
WANG De-bing(Department of Computer and Information Engineering,Anhui Vocational & Technical College of Industry & Trade,Huainan Anhui 232007,China)
出处
《安徽理工大学学报（自然科学版）》
CAS
2018年第5期74-81,共8页
Journal of Anhui University of Science and Technology:Natural Science
基金
安徽省高校自然科学研究基金资助项目(KJ2017A653)
安徽省级质量工程基金资助项目(2016tszy010)
关键词
基准点检测
边缘图构建
基于曲线的角点检测
配准
图像导引神经外科
图像处理
fiducial detection
edge map construction
curve-based corner detection
co-registration
image-guided neurosurgery
image processing
