计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (10): 176-178.

• 图形、图像处理 • 上一篇    下一篇

基于棉纤维截面图像的改进型Canny边缘检测

禹素萍1,3,曾培峰2,陈建平1   

  1. 1.上海交通大学 “区域光纤通信网与新型光通信系统”国家重点实验室,上海 200030
    2.东华大学 计算机学院,上海 201620
    3.东华大学 信息学院,上海 201620
  • 收稿日期:2007-09-29 修回日期:2008-01-02 出版日期:2008-04-01 发布日期:2008-04-01
  • 通讯作者: 禹素萍

Edge detection on fiber cross-sectional image by improved Canny operator

YU Su-ping1,3,ZENG Pei-feng2,CHEN Jian-ping1   

  1. 1.The State Key Lab on Fiber-Optic Local Area Network and Advanced Optical Communication Systems,Shanghai Jiaotong University,Shanghai 200030,China
    2.College of Computer Science and Technology,Donghua University,Shanghai 201620,China
    3.College of Information Science and Technology,Donghua University,Shanghai 201620,China
  • Received:2007-09-29 Revised:2008-01-02 Online:2008-04-01 Published:2008-04-01
  • Contact: YU Su-ping

摘要: 针对棉纤维截面图像的边缘检测,对Canny边缘检测算子提出改进。在边缘跟踪算法中,引入边缘长度L和边缘分叉间平均间隔(像素数)L两个概念,在边缘跟踪算法中进行边缘类型评估。改进后的Canny边缘检测算子与传统Canny边缘检测算子相比,在同等实验条件下,确保了相同棉纤维边缘检出率的同时,更好地抑制了噪声引起的伪边缘输出。

关键词: Canny边缘检测, 纤维截面, 边缘跟踪

Abstract: Edge detection for fiber cross-sectional images is studied in this paper,and the algorithm for improving Canny edge operator is proposed.The edge length L and the average edge phase L are presented to eliminate the fault edge output.With the improved Canny edge operator,the fiber edge output can be improved,and the fault edge output caused by noise can be removed effectively.

Key words: Canny edge detection, fiber cross-section, edge following