Detection of a known target in an image has several different approaches. The complexity and number of steps involved in the target detection process makes a comparison of the different possible algorithm chains desirable. Of the different steps involved, some have a more significant impact than others on the final result the ability to find a target in an image. These more important steps often include atmospheric compensation, noise and dimensionality reduction, background characterization, and detection (matched filtering for this research). A brief overview of the algorithms to be compared for each step will be presented.
This research seeks to identify the most effective set of algorithms for detecting a known tar get. Several different algorithms for each step will be presented, to include ELM, FLAASH, ACORN, MNF, PPI, N-FINDR, MAXD, and two matched filters that employ a structured background model OSP and ASD. The chains generated by these algorithms will be com pared using the Forest Radiance I HYDICE data set. Finally, ROC curves and AFAR values are calculated for each algorithm chain and a comparison of them is presented. Detection rates at a CFAR are also compared. Since a relatively small number of algorithms were used for each step, there were no definitive results generated. However, a comprehensive comparison of the chains using the above mentioned algorithms is presented.
Library of Congress Subject Headings
Remote sensing--Data processing; Image processing--Digital techniques; Computer algorithms
Imaging Science (MS)
Department, Program, or Center
Chester F. Carlson Center for Imaging Science (COS)
Grimm, David C., "Comparison of Hyperspectral Imagery Target Detection Algorithm Chains" (2005). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus