Cluster validity of Xie and Beni and the partition coefficient indexes for fuzzy c-means clustering / Nor Azrin Ahmad Mustaffa

Under Image Processing, there is Image Segmentation. Image Segmentation is a subset of an expansive field of computer vision which deals with partition an image into meaningful regions with respect to a particular application. In particular, it is used to separate regions from the rest of the image,...

Full description

Saved in:
Bibliographic Details
Main Author: Ahmad Mustaffa, Nor Azrin
Format: Thesis
Language:English
Published: 2010
Online Access:https://ir.uitm.edu.my/id/eprint/64309/1/64309.PDF
https://ir.uitm.edu.my/id/eprint/64309/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Under Image Processing, there is Image Segmentation. Image Segmentation is a subset of an expansive field of computer vision which deals with partition an image into meaningful regions with respect to a particular application. In particular, it is used to separate regions from the rest of the image, in order to recognize them as objects. In this project, we implement fuzzy c-means (FCM) clustering which is the technique of segmentation into mammographic images. Segmentation defines the boundary of the targeted object from its background in the images. This project focuses on suspected region that may contain breast anomalies such as masses and calcifications. These breast anomalies may be diagnosed as cancer by radiologists. Therefore, segmentation of mammographic images is an important phase in image analysis that can be further applied to other algorithms for specific tasks such as the detection and classification of breast anomalies. The implementation of FCM for the segmentation of mammographic images is by using Mat lab. FCM is widely used technique in this regard but it requires the priori specification of the number of clusters. Therefore, this project is posed as one of optimization of a fuzzy cluster validity index. There are two validity measures in the context of fuzzy clustering that are being used which are Partition Coefficient and Xie and Beni index. We use C language to write down the cluster validity indexes.