Synthetic Aperture Radar (SAR) image segmentation by fuzzy c-means clustering technique with thresholding for iceberg images

• Published in: Computational ecology and software Vol. 4; no. 2; pp. 129 - 134
• Main Author: Usman Seljuq, Rashid Hussain
• Format: Journal Article
• Language: English
• Published: Hong Kong Computational Ecology and Software 01.06.2014; International Academy of Ecology and Environmental Sciences; International Academy of Ecology and Environmental Sciences (IAEES)
• Subjects: Biological research; Biology, Experimental; Clustering (Computers); Engineering research; Equipment and supplies; fuzzy c-means clustering; Icebergs; Image segmentation; Methods; Properties; Synthetic aperture radar; Synthetic Aperture Radar (SAR); thresholding; Pakistan
• ISSN: 2220-721X; 2220-721X
• DOI: 10.0000/issn-2220-721x-compuecol-2014-v4-0011

Fuzzy c-means (FCM) clustering algorithm is widely used for image segmentation. The purpose of clustering is to identify natural groupings of data from a large data set, which results in concise representation of system's behavior. It can be used to detect icebergs regardless of ambient conditions like rain, darkness and fog. As a result SAR images can be used for iceberg surveillance. In this paper we have investigate FCM with thresholding for iceberg image segmentation for Synthetic Aperture Radar (SAR) images. The results showed that the assessment parameters; mean and entropy have lower values for efficient segmentation.