A multi-scale 3D Otsu thresholding algorithm for medical image segmentation
详细信息 查看全文
- 作者：Yuncong Fenga ; b ; fengyc14@mails.jlu.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Haiying Zhaoc ; zhaohaiying@bupt.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Xiongfei Lia ; b ; lxf@jlu.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Xiaoli Zhanga ; b ; xiaolizhang1987@gmail.com" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Hongpeng Lid ; hongpengli1991@126.com" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：Image segmentation ; 3D Otsu thresholding ; Multi-scale image representation ; Local Laplacian filtering
- 刊名：Digital Signal Processing
- 出版年：2017
- 出版时间：January 2017
- 年：2017
- 卷：60
- 期：Complete
- 页码：186-199
- 全文大小：2789 K
文摘
- •
An efficient 3D Otsu segmentation algorithm is proposed based on dimension decomposition.
- •
Images are smoothed interactively using a fast local Laplacian filter to reduce the influence of noises and weak edges.
- •
A rule for fusing segmentation maps is designed.