A Novel Pixon-Based Image Segmentation Process Using Fuzzy Filtering and Fuzzy C-mean Algorithm

Ehsan Nadernejad, Amin Barari

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Image segmentation, which is an important stage of many image processing algorithms, is the process of partitioning an image into nonintersecting regions, such that each region is homogeneous and the union of no two adjacent regions is homogeneous. This paper presents a novel pixon-based algorithm for image segmentation. The key idea is to create a pixon model by combining fuzzy filtering as a kernel function and a fuzzy c-means clustering algorithm for image segmentation. Use of fuzzy filters reduces noise and slightly smoothes the image. Use of the proposed pixon model prevented image over-segmentation and produced better experimental results than those obtained with other pixon-based algorithms.
Original languageEnglish
JournalInternational Journal of Fuzzy Systems
Volume13
Issue number4
Pages (from-to)350-357
ISSN1562-2479
Publication statusPublished - 2011

Keywords

  • Fuzzy c-mean
  • Image segmentation
  • Clustering
  • Fuzzy filtering
  • Pixonal image

Fingerprint Dive into the research topics of 'A Novel Pixon-Based Image Segmentation Process Using Fuzzy Filtering and Fuzzy C-mean Algorithm'. Together they form a unique fingerprint.

Cite this