{"id":{"repo_id":"mtu","oai_identifier":"oai:digitalcommons.mtu.edu:etds-1967"},"canonical_url":"https://search.ndltd.org/etd/mtu/oai:digitalcommons.mtu.edu:etds-1967","repository":{"repo_id":"mtu","name":"Michigan Technological University","base_url":"https://digitalcommons.mtu.edu/do/oai/"},"display":{"title":"DEEP LEARNING METHODS FOR MULTIBAND EXPLOSIVE HAZARD DETECTION USING L-BAND AND X-BAND FORWARD-LOOKING GROUND-PENETRATING RADAR","abstract":"Explosive hazards are one of the most deadly threats in modern conflicts. The U.S. Army is interested in a reliable way to detect these hazards at range. A promising way of accomplishing this task is using a forward-looking ground-penetrating radar (FLGPR) system. Recently, the Army has been testing a system that utilizes both L-band and X-band radar arrays on a vehicle mounted platform. Using data from this system, we sought to improve the performance of a constant false-alarm-rate (CFAR) prescreener through the use of three deep learning architechtures; deep belief networks (DBNs), stacked denoising autoencoders (SDAEs), and convolutional neural networks (CNNs). We also compare these deep learning classifiers with two more conventional shallow learning classifiers; single kernel support vector machines (SKSVMs) and multiple kernel learning group lasso (MKLGL). By training the deep learners on a combination of image features and comparing the test results to the conventional shallow learners, we were able to significantly increase the probability of detection over both the CFAR prescreener and the shallow learners while maintaining a nominal number of false alarms per square meter. Our research shows that deep learners are a good candidate for improving detection rates in FLGPR systems.","abstract_html":"Explosive hazards are one of the most deadly threats in modern conflicts. The U.S. Army is interested in a reliable way to detect these hazards at range. A promising way of accomplishing this task is using a forward-looking ground-penetrating radar (FLGPR) system. Recently, the Army has been testing a system that utilizes both L-band and X-band radar arrays on a vehicle mounted platform. Using data from this system, we sought to improve the performance of a constant false-alarm-rate (CFAR) prescreener through the use of three deep learning architechtures; deep belief networks (DBNs), stacked denoising autoencoders (SDAEs), and convolutional neural networks (CNNs). We also compare these deep learning classifiers with two more conventional shallow learning classifiers; single kernel support vector machines (SKSVMs) and multiple kernel learning group lasso (MKLGL). By training the deep learners on a combination of image features and comparing the test results to the conventional shallow learners, we were able to significantly increase the probability of detection over both the CFAR prescreener and the shallow learners while maintaining a nominal number of false alarms per square meter. Our research shows that deep learners are a good candidate for improving detection rates in FLGPR systems.","abstract_has_math":false,"creators":["Becker, John T."],"institution":null,"degree_name":"Master of Science in Electrical Engineering (MS)","degree_level":"Master's Thesis","degree_discipline":"Department of Electrical and Computer Engineering","degree_department":null,"school":null,"contributors":["Timothy Havens"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:16:16Z","subjects":["deep learning","explosive hazards","forward-looking","ground-penetrating radar","signal processing","Remote Sensing"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.mtu.edu/etds/968"],"render_values":[{"text":"https://digitalcommons.mtu.edu/etds/968","href":"https://digitalcommons.mtu.edu/etds/968","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/10.37099/mtu.dc.etds/968","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/10.37099/mtu.dc.etds/968","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://digitalcommons.mtu.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.mtu.edu%3Aetds-1967","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Timothy Havens"]},{"key":"dc:creator","label":"Author","values":["Becker, John T."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2015-12-22T08:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2014-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Department of Electrical and Computer Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Master's Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science in Electrical Engineering (MS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["deep learning","explosive hazards","forward-looking","ground-penetrating radar","signal processing","Remote Sensing"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["10.37099/mtu.dc.etds/968","https://digitalcommons.mtu.edu/etds/968"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Explosive hazards are one of the most deadly threats in modern conflicts. The U.S. Army is interested in a reliable way to detect these hazards at range. A promising way of accomplishing this task is using a forward-looking ground-penetrating radar (FLGPR) system. Recently, the Army has been testing a system that utilizes both L-band and X-band radar arrays on a vehicle mounted platform. Using data from this system, we sought to improve the performance of a constant false-alarm-rate (CFAR) prescreener through the use of three deep learning architechtures; deep belief networks (DBNs), stacked denoising autoencoders (SDAEs), and convolutional neural networks (CNNs). We also compare these deep learning classifiers with two more conventional shallow learning classifiers; single kernel support vector machines (SKSVMs) and multiple kernel learning group lasso (MKLGL). By training the deep learners on a combination of image features and comparing the test results to the conventional shallow learners, we were able to significantly increase the probability of detection over both the CFAR prescreener and the shallow learners while maintaining a nominal number of false alarms per square meter. Our research shows that deep learners are a good candidate for improving detection rates in FLGPR systems."]},{"key":"dc:title","label":"Title","values":["DEEP LEARNING METHODS FOR MULTIBAND EXPLOSIVE HAZARD DETECTION USING L-BAND AND X-BAND FORWARD-LOOKING GROUND-PENETRATING RADAR"]}]}],"canonical_facts":{"dc:contributor":["Timothy Havens"],"dc:creator":["Becker, John T."],"dc:date.available":["2015-12-22T08:00:00Z"],"dc:date.created":["2014-01-01T08:00:00Z"],"dc:description.abstract":["Explosive hazards are one of the most deadly threats in modern conflicts. The U.S. Army is interested in a reliable way to detect these hazards at range. A promising way of accomplishing this task is using a forward-looking ground-penetrating radar (FLGPR) system. Recently, the Army has been testing a system that utilizes both L-band and X-band radar arrays on a vehicle mounted platform. Using data from this system, we sought to improve the performance of a constant false-alarm-rate (CFAR) prescreener through the use of three deep learning architechtures; deep belief networks (DBNs), stacked denoising autoencoders (SDAEs), and convolutional neural networks (CNNs). We also compare these deep learning classifiers with two more conventional shallow learning classifiers; single kernel support vector machines (SKSVMs) and multiple kernel learning group lasso (MKLGL). By training the deep learners on a combination of image features and comparing the test results to the conventional shallow learners, we were able to significantly increase the probability of detection over both the CFAR prescreener and the shallow learners while maintaining a nominal number of false alarms per square meter. Our research shows that deep learners are a good candidate for improving detection rates in FLGPR systems."],"dc:identifier":["10.37099/mtu.dc.etds/968","https://digitalcommons.mtu.edu/etds/968"],"dc:subject":["deep learning","explosive hazards","forward-looking","ground-penetrating radar","signal processing","Remote Sensing"],"dc:title":["DEEP LEARNING METHODS FOR MULTIBAND EXPLOSIVE HAZARD DETECTION USING L-BAND AND X-BAND FORWARD-LOOKING GROUND-PENETRATING RADAR"],"thesis:degree_discipline":["Department of Electrical and Computer Engineering"],"thesis:degree_level":["Master's Thesis"],"thesis:degree_name":["Master of Science in Electrical Engineering (MS)"]},"updated_at":"2026-11-19T08:16:16Z"}