摘要
针对梯度幅值边缘检测算法无法检测连续边缘的问题,提出一种自适应多窗口梯度幅值边缘检测算法.首先使用传统梯度幅值边缘检测算法检测出初始边缘;然后在初始边缘上检测端点,使用K-均值聚类算法对端点进行分类,从而确定背景和目标灰度值接近的区域作为窗口;最后在窗口内使用梯度幅值检测边缘,通过多个窗口的并集得到最终的边缘.实验证明采用所提出的算法可以得到比较完整的边缘图,定位误差比传统的边缘检测算法小.
Aiming at the defects of the traditional edge-detection algorithms based on gradient magnitude,which lead to some discontinuous edge,an edge-detection algorithm based on adaptive multi-windows gradient magnitude was proposed.Firstly,traditional gradient magnitude algorithm was adopted to detect the original edge,and then the endpoints in the original edge image were detected.Next,the K-means clustering algorithm was used to classify the endpoints.Finally,the multiple windows using adaptive gradient magnitude were merged to get the final edge map.The experimental results show that the algorithm can get more complete edge map with smaller position error by comparing with the traditional edge detection algorithms.
出处
《华中科技大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2011年第1期14-18,共5页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(50805087
60972162)
湖北省自然科学基金资助项目(2009CDZ027)
三峡大学研究生创新基金资助项目(200914)
关键词
梯度幅值
边缘检测
自适应
多窗口
K-均值聚类
端点检测
gradient magnitude
edge-detection
adaptive
multi-window
K-means clustering
endpoint detection