학술논문
Preliminary study on an improved weight updating for fuzzy c-means with applications to image segmentation
Document Type
Conference
Author
Source
2017 Joint 17th World Congress of International Fuzzy Systems Association and 9th International Conference on Soft Computing and Intelligent Systems (IFSA-SCIS) Fuzzy Systems Association and 9th International Conference on Soft Computing and Intelligent Systems (IFSA-SCIS), 2017 Joint 17th World Congress of International. :1-6 Jun, 2017
Subject
Language
Abstract
Fuzzy c-means is a popular clustering algorithm which allows a single data point to belong to more than one class at any given point. It has been used for a variety of applications especially when the applications are subjective and ambiguous. Image segmentation is one such application in which the decision of a certain pixel belonging to a particular cluster is very fuzzy. The weight associated with every data point is very important as it controls the decision of assigning the data point to a particular cluster. In this study, two novel methods of updating weights that take into account the goodness of clustering and spatial relationships are proposed in order to improve the results of clustering. Fuzzy c-means with our proposed method of updating weights is applied to different kinds of images to perform image segmentation.