• Tipo:
  • Proceedings
Automatic Fuzzy Segmentation of Textural Images Using Adaptive Divergence Affinity Functions

The goal of digital image segmentation is to assign different labels to different objects present in a digital image or volume. A wide variety of sources have been successfully segmented by the traditional fuzzy segmentation algorithm on several different applications. However, the traditional approach of the fuzzy segmentation algorithm usually does not work well when dealing with objects whose materials are characterized by complex textures that cannot be accurately represented by simple fuzzy affinity functions of the traditional fuzzy segmentation algorithm. In this paper, we extend the fuzzy segmentation algorithm to use adaptive textural affinity functions for segmenting these objects in 2D images. These adaptive affinity functions define their optimal appropriate neighborhood size on execution time, by computing texture descriptors surrounding the seed spells, according to the characteristics of the texture being processed. The algorithm then segments the image with the appropriate neighborhood that may be different for each object. We performed experiments on mosaic images that were composed using images from the Brodatz database, and compared our results with the ones produced by a recently published texture segmentation algorithm, showing the applicability of our method.

J. S. Neto, W. Leandro, M. Gadelha, T. Santos, B. M. Carvalho and E. Garduño, "Automatic Fuzzy Segmentation of Textural Images Using Adaptive Divergence Affinity Functions," 2019 International Conference on Systems, Signals and Image Processing (IWSSIP), 2019, pp. 51-56, doi: 10.1109/IWSSIP.2019.8787247.