Toward a Generic Performance Evaluation of Image Segmentation Using Fuzzy C-Means, K-Means and K-Means Using OTSU Thresholding


T.Sam Pradeepraj, Associate Professor, V.Anusuya Devi, Associate Professor
Department of Computer Science and Engineering, School of Computing, Kalasalingam Academy of Research and Education, Krishnankoil, Tamil Nadu, India.


The goal of image segmentation is the splitting of an image into a set of disjoint regions with uniform and homogeneous attributes such as intensity, color, tone or texture etc. In many real locales, for images, issues such as limited spatial resolution, indigent contrast, overlapping intensities, noise and intensity inhomogeneities introduce fuzziness in the object boundaries in the image. Due to this the fuzzy set theory was proposed. This work consists of three segmentation algorithm such as fuzzy c-means, k-means and k-means using Otsu thresholding. The proposed work's aim is to evaluate the performance of above three image segmentation algorithms on synthetic images. Experiments show that the K means using Otsu thresholding algorithm effectively segments the given image other than two algorithms.