Fuzzy Clustering with Multi-Resolution Bilateral Filtering for Medical Image Segmentation

Fuzzy Clustering with Multi-Resolution Bilateral Filtering for Medical Image Segmentation

Kai Xiao (School of Software, Shanghai Jiaotong University, Shanghai, China), Jianli Li (School of Software, Shanghai Jiaotong University, Shanghai, China), Shuangjiu Xiao (School of Software, Shanghai Jiaotong University, Shanghai, China), Haibing Guan (Shanghai Key Laboratory of Scalable Computing and Systems, Department of Computer Science, Shanghai Jiaotong University, Shanghai, China), Fang Fang (Shanghai First People’s Hospital, Shanghai, China) and Aboul Ella Hassanien (Department of Information Technology, Cairo University, Giza, Cairo, Egypt)
Copyright: © 2013 |Pages: 13
DOI: 10.4018/ijfsa.2013100104
OnDemand PDF Download:
List Price: $37.50
10% Discount:-$3.75


Although fuzzy c-means (FCM) algorithm and some of its variants have been extensively widely used in unsupervised medical image segmentation applications in recent years, they more or less suffer from either noise sensitivity or loss of details, which always is a key point to medical image processing. This paper presents a novel FCM variation method that is suitable for medical image segmentation. The proposed method, typically by incorporating multi-resolution bilateral filter which is combined with wavelet thresholding, provides the following advantages: (1) it is less sensitive to both high- and low-frequency noise and removes spurious blobs and noisy spots, (2) it yields more homogeneous clustering regions, and (3) it preserves detail, thus significantly improving clustering performance. By the use of synthetic and multiple-feature magnetic resonance (MR) image data, the experimental results and quantitative analyses suggest that, compared to other fuzzy clustering algorithms, the proposed method further enhances the robustness to noisy images and capacity of detail preservation.
Article Preview

Image segmentation is an important processing step in many image, video and computer vision applications. Extensive research has been done in creating many different approaches and algorithms for image segmentation, but it is still difficult to assess whether one algorithm produces more accurate segmentations than another, whether it be for a particular image or set of images, or more generally, for a whole class of images (Hui et al. 2008).

Generally, the pixels on an image are correlated at certain degree, i.e. the pixels in the immediate neighbors possess relatively similar feature data. In other words, normally the probability that adjacent pixels belong to the same cluster will be high. This is especially true in medical images. Therefore, taking advantage of this spatial information can assist medical image segmentation.

Complete Article List

Search this Journal:
Volume 12: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 11: 4 Issues (2022)
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing