Density Based Fuzzy Thresholding for Image Segmentation

Citation:
Li, J., B. Dai, K. Xiao, and A. E. Hassanien, "Density Based Fuzzy Thresholding for Image Segmentation", Advanced Machine Learning Technologies and Applications (AMLTA), Cairo Egypt, pp. 118--127, 2012.

Abstract:

In this paper, we introduce an image segmentation framework which
applies automatic threshoding selection using fuzzy set theory and fuzzy
density model. With the use of different types of fuzzy membership function,
the proposed segmentation method in the framework is applicable for images of
unimodal, bimodal and multimodal histograms. The advantages of the method
are as follows: (1) the threshoding value is automatically retrieved thus requires
no prior knowledge of the image; (2) it is not based on the minimization of a
criterion function therefore is suitable for image intensity values distributed
gradually, for example, medical images; (3) it overcomes the problem of local
minima in the conventional methods. The experimental results have
demonstrated desired performance and effectiveness of the proposed approach.

PreviewAttachmentSize
3220118.pdf2.89 MB