A local Zernike moment-based unbiased nonlocal means fuzzy C-Means algorithm for segmentation of brain magnetic resonance images

作者：

Highlights：

• A nonlocal and local transform-based method is proposed for MR image segmentation.

• Local Zernike moments (LZMs) are used to reduce Rician noise from MR images.

• Fuzzy C-means is used to represent nonlocal and LZMs for local information.

• A complete framework of the proposed method is explained.

• Superior image segmentation performance is demonstrated through detailed experimental analysis.

摘要

•A nonlocal and local transform-based method is proposed for MR image segmentation.•Local Zernike moments (LZMs) are used to reduce Rician noise from MR images.•Fuzzy C-means is used to represent nonlocal and LZMs for local information.•A complete framework of the proposed method is explained.•Superior image segmentation performance is demonstrated through detailed experimental analysis.

论文关键词：MR image segmentation,Zernike moments,Local Zernike moments,Fuzzy C-means,Rician noise

论文评审过程：Received 2 April 2018, Revised 24 September 2018, Accepted 13 October 2018, Available online 14 October 2018, Version of Record 25 October 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.10.023