Flat EEG Image Segmentation by Fuzzy Entropy-Based Multi-Level Thresholding

Abstract

Thresholding is a type of image segmentation that deals with the conversion of an image with many gray levels into another image with fewer gray levels. It classifies grayscale pixels into two categories which creates a binary image. However, the output image is not always
satisfying due to several factors such as inherent image vagueness as uncertainty arises within the gray values of an image. In this paper, a multi-level image thresholding based on fuzzy entropy is applied on Flat Electroencephalography (Flat EEG) image. The outcomes are compared visually with global thresholding.