摘要
针对形态学单一结构元素在边缘检测中图像边缘信息丢失的问题,根据图像形态学多刻度形态滤波的思想,提出了基于顺序形态学自适应边缘检测方法。根据传统的形态学边缘检测方法,该方法构造了边缘检测算子,给出了边缘检测算子的评价函数。通过该函数选择不同方向的结构元素增加图像的边缘信息。实验结果表明,与传统的基于数学形态学的边缘检测方法相比,该方法可以在最大程度上抑制噪声,有效地提高图像的边缘检测效果。
In order to resolve the image edge missing problem of traditional edge detecting method,this paper proposed an adaptive order morphology detecting method based on the theory of morphological filter.In order to enhance the image edge by selecting different direction＇s structure element,the method constructed a evaluation function based on traditional morphological edge detecting method.Experiment results show that,compared with the traditional edge detecting method,this method decreases the noise as possible as it can and enhances image edge.
出处
《计算机应用研究》
CSCD
北大核心
2011年第5期1978-1980,共3页
Application Research of Computers
关键词
数学形态学
边缘检测
结构元素
噪声抑制
mathematical morphological transform
edge detecting
structural element
noise restraining
