Arnab Shaw (Advisor)
Master of Science in Engineering (MSEgr)
This thesis studies the detection of changes using hyperspectral images. Change Detection (CD) is the process of identifying and examining temporal and spectral changes in signals. Detection and analysis of change provide valuable information of possible transformations in a scene. Hyperspectral imaging (HSI) sensors are capable of collecting data at hundreds of narrow spectral bands. Such sensors provide high-resolution spatial and spectrally rich information that can be exploited for CD. This work develops and implements various CD algorithms for detection of changes using Hyperspectral images. The main objectives are to study and develop different HSI change detection algorithms. The explored methods were implemented in order to compare the performance on close-in HSI data. The methods studied in this thesis include, Image Differencing, Image Ratioing, Principal Component Analysis, Linear Chronochrome, a modified Correlation Coefficient and a Kernel Dissimilarity Measure. Hyperspectral imagery of different scenarios was collected and used to test and validate the methods presented in this study. The algorithms were implemented using MATLAB, and the performance of algorithms is presented in terms of false alarm rates and missed changes.
Department or Program
Department of Electrical Engineering
Year Degree Awarded
Copyright 2007, all rights reserved. This open access ETD is published by Wright State University and OhioLINK.