交互式数字图象的配准与镶嵌主要由二个步骤来完成:第一步是由人在显示屏幕上选择确定图像几何位置的控制点集,然后由此确定一幅图像几何变换使两幅图像在几何位置上配准。
Step one is to pick up the control point set which determines the geometric position of images, and then to derive a geometric transform for registering one image to another from those points.
实验结果表明在该软件系统中采用基于控制点的配准方法可有效解决图像配准问题,而且具有较高的配准精度。
The experiments shows that image registration method based on control point is effective in image registration, and it has higher registration precision.
该方法在基于控制点的多项式粗配准的基础上,利用分类图像相关实现了基于仿射变换模型的精配准。
A feature matching (algorithm based) on classification is proposed, which uses the correlation of the classified images to realize a precise registration after a coarse (registration).
这里提出一种基于控制点搜索的最大互相关算法来完成配准任务。
Here, we proposed an algorithm which based on the control point to search the biggest mutual correlation to complete the registration.
这里提出一种基于控制点搜索的最大互相关算法来完成配准任务。
Here, we proposed an algorithm which based on the control point to search the biggest mutual correlation to complete the registration.
应用推荐