摘要
由于实际应用中 ,仅有边缘轮廓信息并不足以准确、有效地描述被检测物体 ,所以如何根据边缘的方向链码来重建二值图象是图象处理应用中的一类常见问题 .人们已经研究过许多不同的方法 ,例如奇偶检测法、种子法等等 .但这些方法或者由于计算量大 ,或者可能出现判别错误的现象 ,因而不能满足实际需求 .为了准确、快速地解决这一类问题 ,因此在方向链码的基础上 ,先将边缘点按照区域填充法则进行分类 ,进而提出了一种依据方向链码和边缘点分类 ,来重建二值图象的新方法 .实践证明 ,该方法可以很好地克服其它算法的缺点 ,并可得到较好的检测效果 ,而且具有准确、快速、计算量小、无漏检等优点 .现已被应用于作者所开发的“电力工业用射线探伤辅助识别系统”中 .
Edge extraction is an usual method in the field of image processing. But it's not enough to exactly describe the detected object only using edge information in the practical applications. So how to restore the binary image according to the direction chain codes is an important problem in image processing field. A binary image can be described by using one or several direction chain codes that record the contour of this binary image. Many methods heve been proposed in the literature, but there still exist the drawbacks such as large computation and inaccurate restoration. In this paper, a new and fast method for restoration of binary image is presented to solve the problems, which is based on the sort of edge points and the chain coding of the boundaries. Experimental results show that this method is computationally faster than the previous methods and does not miss any part of the interior. It's a good restoration method, and already has been used in the practical application.
出处
《中国图象图形学报（A辑）》
CSCD
2000年第6期474-478,共5页
Journal of Image and Graphics
基金
国家自然科学基金!( 69672 0 14 )
国家"九五"重点科技项目
