J4 ›› 2010, Vol. 48 ›› Issue (05): 835-839.

• 计算机科学 • 上一篇    下一篇

喷丝板单孔图像的检测算法

张佳全1, 吴庆妍2, 冯玮琛3   

  1. 1. 吉林大学 物理学院, |长春 130012|2. 吉林大学 公共计算机教学与研究中心, 长春 130022;3. 吉林大学 数学学院, 长春 130012
  • 收稿日期:2010-03-23 出版日期:2010-09-26 发布日期:2010-09-21
  • 通讯作者: 吴庆妍 E-mail:wuqy@jlu.edu.cn

Detecting Algorithm |of |Spinneret Hole Image

ZHANG Jia quan1, WU Qing yan2, FENG Wei chen3   

  1. 1. College of Physics, Jilin University, Changchun 130012, China;
    2. Center for Computer Fundamental Education, Jilin University, Changchun 130022, China;
    3. College of Mathematics, Jilin University, Changchun 130012, China
  • Received:2010-03-23 Online:2010-09-26 Published:2010-09-21
  • Contact: WU Qing yan E-mail:wuqy@jlu.edu.cn

摘要:

运用计算机图像处理方法检测化纤喷丝板单孔孔径和堵塞. 使用二维转台和CCD摄像头采集喷丝板单孔图像, 并提出一种新的图像检测算法: 先对喷丝孔图像进行傅里叶变换, 再采用Sobel算子边缘检测和Sobel算子边缘细化法,  依据圆孔几何特征判定喷丝孔是否堵塞, 依据其图像所占像素量确定堵塞面积比, 并求得未堵塞喷丝孔的直径.  实际应用表明, 所提算法合理、 有效, 达到了一定的测量精度.  

关键词: 图像处理, 喷丝孔, 边缘检测, 算法, 圆孔识别, 面积比

Abstract:

The occluded spinneret hole and the diameter of spinneret were detected by means of computer image processing. Meanwhile, an algorithm was introduced to detect the spinneret hole image which was collected by a CCD camera, including image collection, square Gray-scale transformation, edge detection based on Sobel algorithm, binarization based on distance -preserving method, an edge thinning algorithm with Sobel operator. On the basis of geometrical characteristics of the circle, we judged  if spinneret holes had been occluded, and determined clogging area proportion via  the pixel of the spinneret hole image and solved the diameter of the hole that had not been occluded. The application  indicates that this algorithm is reasonable, efficient and accurate.

Key words: image processing, spinneret hole, edge detection, algorithms, hole identification, area proportion

中图分类号: 

  • TP391.41