摘要
图像匹配是图形图像处理领域研究的重要问题之一 .针对基于层析图像的逆向工程 CAD建模过程中的图像轮廓识别与匹配问题 ,通过建立物体轮廓森林 ,进行轮廓识别 ,缩小轮廓匹配搜索范围 ;利用椭圆定量表达平面离散点序列的整体形位信息 ,实现轮廓匹配 ;提出加权旋转半分法处理分枝问题 ;并结合典型算例 ,进行算法验证 ,取得了较好效果 .本算法具有简单易用、效率高的优点 .
Contour matching is one of the important problems concerned in graph and image processing field. Aiming at implementing the planar contour recognizing and matching in the CAD modeling based on the slice images, some fruitful research works are presented in this paper. Firstly, a contour forest is established to reduce the searching range in the contour matching and to recognize the in-out features of contours. An ellipse-fitting algorithm is then applied to describe the completed shape and locating information for the series of discrete contour points and to realize primarily the slice contour matching. Finally, a weighted rotating-branching algorithm has been proposed to resolve the branching problem. The efficiency of this algorithm has been verified by several typical examples and some satisfying results have been obtained.
出处
《计算机研究与发展》
EI
CSCD
北大核心
2002年第9期1127-1131,共5页
Journal of Computer Research and Development
基金
航空基础科学基金资助 ( 99H5 3111)
关键词
产品层析图像
轮廓序列匹配
逆向工程
椭圆拟合
加权旋转半分法
图形图像处理
CAD
reverse engineering, slice images, contour recognition, contour matching, contour forest, ellipse-fitting, branching, weighted rotating-branching algorithm
