摘要
图像的细化是提取曲线特征,矢量化、识别等后续图像处理的基础,而单一的算法往往不能得到理想的细化效果。本文在研究各种已有的细化算法的基础上,从中选出两种不同的数学形态学算法共同处理同一图像,可以实现较好的细化效果:即先对图像采用基于5×5邻域模板的数学形态学细化算法进行细化处理,再对处理后的图像采用基于结构元素对的数学形态学细化方法处理。通过编程和各种方法相比较,证明采用这种方式可以实现无腐蚀,连通性好,无毛刺,且为八邻连接的细化效果,为后续的图像处理打下良好的基础。
Image thinning is the foundation of the follow-up image processing such as curve feature extraction, vectorization, identification etc. However a single thinning algorithm often cannot reach ideal thinning effects. The paper proposes a method of image thinning based on composite algorithm of mathematical morphology： first, thinning processing based on mathematical morphology method of 5 × 5 neighborhood template; second, thinning processing based on mathematical morphological method of structure elements. The experiment results indicate that the method can realize non-corrosive, good connectivity, without burr and for eight adjacent connections effect, and lay a good foundation for the follow-up of image processing.
出处
《自动化技术与应用》
2011年第10期67-71,共5页
Techniques of Automation and Applications
关键词
数学形态学
图像
细化
Mathematical morphology
Image
Thinning
