Interest-based ordering for fuzzy morphology on white blood cell image segmentation

Chastine Fatichah, Martin Leonard Tangel, Muhammad Rahmat Widyanto, Fangyan Dong, Kaoru Hirota

Research output: Contribution to journalArticlepeer-review

28 Citations (Scopus)


An Interest-based Ordering Scheme (IOS) for fuzzy morphology on White-Blood-Cell (WBC) image segmentation is proposed to improve accuracy of segmentation. The proposed method shows a high accuracy in segmenting both high- and low-density nuclei. Further, its running time is low, so it can be used for real applications. To evaluate the performance of the proposed method, 100 WBC images and 10 leukemia images are used, and the experimental results show that the proposed IOS segments a nucleus in WBC images 3.99% more accurately on average than the Lexicographical Ordering Scheme (LOS) does and 5.29% more accurately on average than the combined Fuzzy Clustering and Binary Morphology (FCBM) method does. The proposal method segments a cytoplasm 20.72% more accurately on average than the FCBM method. The WBC image segmentation is a part of WBC classification in an automatic cancer-diagnosis application that is being developed. In addition, the proposed method can be used to segment any images that focus on the important color of an object of interest.

Original languageEnglish
Pages (from-to)76-86
Number of pages11
JournalJournal of Advanced Computational Intelligence and Intelligent Informatics
Issue number1
Publication statusPublished - Jan 2012


  • Binary morphology
  • Color ordering scheme
  • Fuzzy morphology
  • Image segmentation
  • White-blood-cell image


Dive into the research topics of 'Interest-based ordering for fuzzy morphology on white blood cell image segmentation'. Together they form a unique fingerprint.

Cite this