摘要
本文从图论的观点出发 ,通过抽取细化后图像中的端点、歧点和叉点等特征点作为图的顶点 ,计算出图的边数ε和顶点数v ,在分析二者数量关系的基础上 ,确定图像中闭合曲线的存在性 .本方法实现简单 ,抗噪能力强 ,是一种实用高效的闭合曲线检测方法 .
A thinned image is viewed as a graph,and feature points such as ending,bifurcation and crossover are extracted as the vertexes.By counting the number of vertexes and Sides and analyzing their quantitative relationship,whether closed curves exist or not in an image is determined.The proposed method is easy to implement and has strong anti-noise ability.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2002年第8期1232-1234,共3页
Acta Electronica Sinica
基金
宁波市青年基金 (No 0 2J2 0 1 0 2 - 0 2 )
浙江省综合信息网技术重点实验室项目 (No 1 1 1 4 0 2 -J30 1 74)
关键词
特征点
闭合曲线
图像处理
端点
歧点
叉点
connected graph
feature point
closed curve
image processing
structural analysis
