Land Cover Segregation Of Coastal Area Using K Means Algorithm

2137 Words Jan 13th, 2015 9 Pages
Land-Cover Segregation of Coastal Area using K-Means Algorithm

Manjari Saha
Computer Science and Engg. Dept.,
Govt. College of Engineering and Textile Technology, Serampore,
Hooghly, West Bengal, PIN-712201. India.
E-Mail: cmanjari@gmail.com

Abstract
Coastal areas provide livelihood facilities to many and also offer vast recreational and economic activities, and yet, at the same time, such eco-regions require to be managed with special emphasis and consideration of its natural and cultural resources. Land cover classification of such diverse sea-land regions requires a high level of effort and plays a vital role for the analysis of time-based or event-based change on certain areas. For effective utilization of remote sensed images obtained from Earth-orbiting satellites, a lot of image classification methods are available in literature. In this paper, we use the method of unsupervised classification for the users ' convenience and flexibility, low computational requirements and moderate classification accuracy. The objectives of the paper is to classify the land cover by mapping the region into W-S-V (Water, Soil, Vegetation) components using the unsupervised K-means algorithm, to obtain land- use/cover gray scale image combining the W-S-V components and thereby finally deriving the related performance metrics such as histogram and land cover correlation.

Keywords - Clustering, Coastal area, Correlation, Histogram, Image segmentation, K-Means, Land Cover/Use…
Open Document