Shoreline Change Detection: Methodology

Shoreline change detection: Methodology
Student:
Professor:
Course title:
Date:
Shoreline change detection: Methodology
Methodology
To detect and measure changes in the coastline, this research will follow a three-step methodology. The first step will entail preprocessing in order to automatically take out the study area from satellite pictures. In the second step, NDWI index from Landsat TM pictures will be utilized after binarization in order to detect coastlines. It is worth mentioning that NDWI is defined in the following way: NDWI = (Green – NIR) / (Green + NIR). NIR is band 4 whereas Landsat TM Green is band 2. This index is designed to make the most of reflectance of water with the use of green wavelengths; to reduce the low reflectance of NIR by water features; and to maximize high reflectance of NIR by soil and vegetation features. Consequently, soil and vegetation have negative values or zero and are suppressed, whereas water features have positive values are therefore enhanced. The third and final step is essentially the measurement of changes as well as the prediction of the future alters.
Geometric correction of satellite images
The geometric correction will be carried out with the use of 45 in situ Ground Control Points. Street intersections situated throughout the image will be selected as Ground Control Points and the correction will use a linear model that involves coordinate translation. Thorough visual inspection will be performed to ensure that the images actually match the Ground Control Points.
Preprocessing
The input for measuring changes will be manually cropped image of the study area. The image will be utilized as reference in extracting the study area from several pictures using Speeded Up Robust Features (S...