"Landmine detection using hand-held units is a difficult problem due to varying type and composition of metals in landmines. This research explores spatially distributed features to discriminate landmines from other harmless objects. The feature calculation involves the wavelet decomposition of Metal Detector (MD) energy sequences to obtain the approximate and detailed coefficients, and correlation of these coefficients with Weighted Density Distribution (WDD) functions. The features calculated are evaluated on a standard back propagation neural network on real data sets with more than 1500 mine encounters of varying shape, size and metal content. The effectiveness of wavelet decomposition and WDD functions is investigated"--Abstract, page iii.
Stanley, R. Joe
Agarwal, Sanjeev, 1971-
Moss, Randy Hays, 1953-
Electrical and Computer Engineering
M.S. in Electrical Engineering
University of Missouri--Rolla
viii, 50 pages
© 2004 Kalyan Ram Achanta, All rights reserved.
Thesis - Restricted Access
Land mines -- Detection
Mines (Military explosives) -- Detection
Print OCLC #
Link to Catalog Record
Electronic access to the full-text of this document is restricted to Missouri S&T users. Otherwise, request this publication directly from Missouri S&T Library or contact your local library.http://merlin.lib.umsystem.edu/record=b5382401~S5
Achanta, Kalyan Ram, "Landmine discrimination using wavelet decomposition and weighted density distribution functions" (2004). Masters Theses. 3621.
Share My Thesis If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the button above.