Noise robust image edge detection based upon the automatic anisotropic Gaussian kernels

详细信息    查看全文

• 作者：WeiChuan Zhang^a ; ^b ; ^{zwc2003@163.com} ; ^{weichuan.zhang@durham.ac.uk}Author Vitae ; YaLi Zhao^aAuthor Vitae ; Toby P. Breckon^bAuthor Vitae ; Long Chen^c
• 关键词：Automatic anisotropic Gaussian kernels ; Anisotropic directional derivatives (ANDDs) ; Edge detection ; Canny detector
• 刊名：Pattern Recognition
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：63
• 期：Complete
• 页码：193-205
• 全文大小：4838 K
• 卷排序：63

文摘

The reasons why some obvious edge miss detected in Canny detector are analyzed thoroughly. The automatic anisotropic Gaussian kernels are designed and a revised edge extraction is proposed. A noise robust edge detector with excellent performance is presented.