Universiti Teknologi Malaysia Institutional Repository

Flood detection of tsunami affected areas using multispecteral images

Siti Yuhaniz, Siti Yuhaniz (2008) Flood detection of tsunami affected areas using multispecteral images. In: Asian Association on Remote Sensing (AARS) - Proceeding, 2008, Sri Lanka.

[img]
Preview
PDF
68Kb

Abstract

The Disaster Monitoring Constellation (DMC) is a constellation of Earth observing small satellites developed by the Surrey Satellite Technology Limited (SSTL) and jointly owned by several countries. The DMC offers a novel approach to disaster monitoring by providing daily coverage of any place of the Earth using medium spatial resolution multispectral images. The DMC satellites carry multispectral imagers that have three bands, which are similar to the bands of the Landsat imagers. This paper presents a study on flood detection methods using images captured by the DMC satellites of the Southeast Asia Tsunami disaster that happened last year. The research is aimed at automating the flood detection process and finding the most optimal method, in term of accuracy and computing resources used, so that it is suitable for implementation on-board small satellites. The potential use of DMC images for detecting flooding areas is tested. The NDVI differencing and Near-infrared band differencing and four conventional classification methods (parallelepiped, maximum likelihood, minimum distance and mahalanobis distance) are investigated for their suitability to detect flooding areas. The efficiency of the investigated flood detection methods is compared based on DMC multispectral data of Tsunami affected areas and reference pre-flooding images from Landsat. Testing results illustrating their accuracy as well as the performance and required resources of the on-board computer systems are presented.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:flood detection, change detection, on-board processing
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:Computer Science and Information System (Formerly known)
ID Code:19755
Deposited By: Liza Porijo
Last Modified:15 Dec 2011 07:53

Repository Staff Only: item control page