Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (30): 178-180.DOI: 10.3778/j.issn.1002-8331.2010.30.052

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

New image edge detection method on nonsubsampled Contourlet transform

LI Xing-mei1,2,YAN Guo-ping1   

  1. 1.Department of Electronics and Information,Huazhong University of Science and Technology,Wuhan 430074,China
    2.Faculty of Mechanical and Electronic Information,China University of Geosciences,Wuhan 430074,China
  • Received:2009-03-16 Revised:2009-05-15 Online:2010-10-21 Published:2010-10-21
  • Contact: LI Xing-mei

无下采样Contourlet变换在图像边缘检测中的应用

李杏梅1,2,严国萍1   

  1. 1.华中科技大学 电信系,武汉 430074
    2中国地质大学 机电学院,武汉 430074
  • 通讯作者: 李杏梅

Abstract: Contourlet can decompose the image in multi-scale and multi-direction,which has better effect on presentation the edge.Though Contourlet can be applied in many fields in image,there are little articles about the application in edge detection.Based on the idea of the model of anisotropic receptive fields which can be applied in high pass filtering,the nonsubsampled Contourlet transform is proposed to be used in edge detection.The result shows the method is effective.

Key words: Contourlet transform, image edge detection, the model of anisotropic receptive fields

摘要: 传统图像边缘检测不能同时实现边缘检测需要的各向异性和多尺度性,小波虽然可以做到,但是小波在表现多方向性时,不能以最稀疏的方式表示。Contourlet变换正是解决这些问题的一种新的分析工具。目前将Contourlet变换用于图像边缘检测的方法还很少见,该文在各向异性的感受野模型可以很好用于图像高通滤波的思想上,提出一种利用无下采样Contourlet变换进行图像边缘检测的方法。实验结果证明,该方法可以较好地用于图像的边缘检测。

关键词: Contourlet变换, 图像边缘检测, 各向异性的感受野模型

CLC Number: