High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

An Efficient Detection of Flood Extent from Satellite Images using Contextual Features and Optimized Classification

Author(s):

Prabhaa N M , JAI SHRI RAM GROUP OF INSTITUTIONS; J. Nanthini, JAI SHRI RAM GROUP OF INSTITUTIONS; Sakthivel K., NACHIMUTHU POLYTECHNIC COLLEGE,ECE

Keywords:

Flood Mapping, GIS, TerraSAR-X, Landsat, Remote Sensing, Rule-based Classification, Confusion Matrix

Abstract

Floods are among the most destructive natural disasters worldwide. In flood disaster management programs, flood mapping is an initial step. It provides an efficient methodology to recognize and map flooded areas by using TerraSAR-X imagery. First, a TerraSAR-X satellite image is captured during a flood event in some places, to map the inundated areas. Multispectral Landsat imagery is then used to detect water bodies prior to the flooding. Both the two terrasar-x image and landsat image are compared according to the three parameters such as shape, color and scale etc. Before that the speckles are removed from both the images using frost filter then gaps are filled using Local linear histogram matching method after that pan sharpening is done by gram-schmidt spectral sharpening method. The second stage is segmentation of the images. Segmentation is done by multi-resolution process using Taguchi technique. In synthetic aperture radar (SAR) imagery, the water bodies and flood locations appear in black; thus, both objects are classified as one. The class of the water bodies was extracted from the Landsat image and then subtracted from that extracted from the TerraSAR-X image. The remaining water bodies represented the flooded locations. Object-oriented classification and Taguchi method are implemented for both images. The Landsat images are categorized into three classes, namely, urban, vegetation, and water bodies. The classification results are evaluated using a confusion matrix. Consequently, the flooded locations are recognized and mapped by subtracting the two classes of water bodies from these images.

Other Details

Paper ID: IJSRDV4I110043
Published in: Volume : 4, Issue : 11
Publication Date: 01/02/2017
Page(s): 102-106

Article Preview

Download Article