A novel image thresholding method based on Parzen window estimate

Shitong Wang, Fu Lai Korris Chung, Fusong Xiong

Research output: Journal article publicationJournal articleAcademic researchpeer-review

111 Citations (Scopus)

Abstract

Image segmentation is one of the most important and fundamental tasks in image processing and techniques based on image thresholding are typically simple and computationally efficient. However, the image segmentation results depend heavily on the chosen image thresholding methods. In this paper, histogram is integrated with the Parzen window technique to estimate the spatial probability distribution of gray-level image values, and a novel criterion function is designed. By optimizing the criterion function, an optimal global threshold is obtained. The experimental results for synthetic real-world and images demonstrate the success of the proposed image thresholding method, as compared with the OTSU method, the MET method and the entropy-based method.
Original languageEnglish
Pages (from-to)117-129
Number of pages13
JournalPattern Recognition
Volume41
Issue number1
DOIs
Publication statusPublished - 1 Jan 2008

Keywords

  • Image segmentation
  • Parzen window
  • Thresholding

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this