Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (2): 153-155.DOI: 10.3778/j.issn.1002-8331.2011.02.047

• 图形、图像、模式识别 • Previous Articles     Next Articles

Crack edge detection using comprehensive CV and facet model

XIANG Caibing1,2,ZENG Li1,2   

  1. 1.College of Mathematics and Physics,Chongqing University,Chongqing 400044,China
    2.ICT Research Center,Key Laboratory of Optoelectronic Technology and System of the Education Ministry of China,Chongqing University,Chongqing 400044,China
  • Received:2009-08-25 Revised:2010-01-05 Online:2011-01-11 Published:2011-01-11
  • Contact: XIANG Caibing

综合CV和Facet模型的裂纹边缘检测

向才兵1,2,曾 理1,2   

  1. 1.重庆大学 数理学院,重庆 400044
    2.重庆大学 光电技术及系统教育部重点实验室 ICT研究中心,重庆 400044
  • 通讯作者: 向才兵

Abstract: Edge detection of CV model is continuous and closed,but the accuracy is not high.While the result of Facet model has high accuracy,but is not continuous.A new method of comprehensive CV model and Facet model is researched in this paper.This method includes several parts as follow.Firstly,the crack edge is detected by CV model in the whole image.Secondly,the new crack edge is detected by Facet model only in the neighborhood of CV detection result.Thirdly,the final crack edge is attained by combining detection results of CV and Facet together.This method is applied to crack edge detection of industrial CT images.The experimental results show the effectiveness of it.

Key words: edge detection, CV model, Facet model, crack detection

摘要: CV模型能检测到连续封闭的裂纹边缘,但边缘的定位精度不高,Facet模型定位裂纹边缘的精度高,但不连续。针对上述问题,研究了一种综合CV和Facet模型的算法。该算法利用CV模型进行边缘检测,在CV边缘点的附近运用Facet模型进行检测,将两种方法分别获得的边缘点进行融合得到连续封闭且定位准确的边缘。将该算法应用于工业CT图像的裂纹边缘检测中,实验结果表明该方法是有效的。

关键词: 边缘检测, CV模型, Facet模型, 裂纹检测

CLC Number: