{"689059":{"#nid":"689059","#data":{"type":"event","title":"PhD Defense by HANA HERNDON","body":[{"value":"\u003Cp\u003ESchool of Civil and Environmental Engineering\u003C\/p\u003E\u003Cp\u003EPh.D. Thesis Defense Announcement\u003C\/p\u003E\u003Cp\u003EENHANCING STRUCTURAL INSPECTIONS BY INTEGRATING COMPUTER VISION, MACHINE LEARNING, AND RISK ASSESSMENT FOR COMPREHENSIVE CORROSION EVALUATION\u003C\/p\u003E\u003Cp\u003EBy HANA HERNDON\u003C\/p\u003E\u003Cp\u003EAdvisors:\u003C\/p\u003E\u003Cp\u003EDR. IRIS TIEN\u003C\/p\u003E\u003Cp\u003EDR. DAVID FROST\u003C\/p\u003E\u003Cp\u003ECommittee Members:\u003C\/p\u003E\u003Cp\u003EDR. MAHDI ROOZBAHANI (GT\/CSE)\u003C\/p\u003E\u003Cp\u003EDR. DONALD WHITE (GT\/CEE)\u003C\/p\u003E\u003Cp\u003EDR. JENNIFER MCCONNELL (UDEL\/CCEE)\u003C\/p\u003E\u003Cp\u003EDate and Time: Wednesday, April 15, 2026, 12PM\u003C\/p\u003E\u003Cp\u003ELocation: Pettit 102A\/ MS Teams Meeting\u003C\/p\u003E\u003Cp\u003EABSTRACT\u003Cbr\u003ECorrosion causes irreversible damage to structures, and understanding its extent\u003Cbr\u003Eand impact is crucial for engineers to prioritize repairs and prevent failures.\u003Cbr\u003EHowever, traditional bridge inspections can be inefficient, costly, and unsafe.\u003Cbr\u003EUnmanned aerial vehicles (UAVs) are revolutionizing bridge inspections by enabling\u003Cbr\u003Erapid, remote data collection. But without rigorous data-processing methodologies, UAV-aided inspections still require extensive manual post-processing. This research\u003Cbr\u003Edevelops a computational framework to assess corrosion on bridges from images,\u003Cbr\u003Equantify its structural impact, and translate those findings into assessments of\u003Cbr\u003Estructural risk, improving inspection efficiency and effectiveness. Three research\u003Cbr\u003Eefforts support this objective. First, a computer vision methodology was developed\u003Cbr\u003Eto segment corrosion in UAV images of Georgia bridges. Results show a 30%\u003Cbr\u003Eimprovement in recall compared to existing methods, reducing the likelihood of\u003Cbr\u003Emissed damage. Next, a technique to assess corrosion severity and predict section\u003Cbr\u003Eloss from images was developed. Steel coupons underwent accelerated corrosion\u003Cbr\u003Etesting with mass loss recorded and images captured over time. Image features\u003Cbr\u003Ewere used to train tree-based ensemble models, and a Random Forest model\u003Cbr\u003Eachieved a normalized root mean square error of 0.12, demonstrating reliable\u003Cbr\u003Eestimation of section loss from images alone. Finally, image-derived damage data\u003Cbr\u003Ewere integrated into stochastic models to evaluate residual capacity and reliability.\u003Cbr\u003EMonte Carlo simulations and variance-based sensitivity analyses quantified\u003Cbr\u003Euncertainty and identified influential parameters, showing that image-based inputs\u003Cbr\u003Econtribute comparably to traditional structural variables. Together, this work\u003Cbr\u003Eadvances remote structural inspections and supports data-driven infrastructure\u003Cbr\u003Edecision-making.\u003Cbr\u003EFull link:\u003Cbr\u003Ehttps:\/\/teams.microsoft.com\/meet\/24453127545336?p=BaI5fjK3bovb1vAY1U\u003C\/p\u003E","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EENHANCING STRUCTURAL INSPECTIONS BY INTEGRATING COMPUTER VISION, MACHINE LEARNING, AND RISK ASSESSMENT FOR COMPREHENSIVE CORROSION EVALUATION\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":[{"value":"ENHANCING STRUCTURAL INSPECTIONS BY INTEGRATING COMPUTER VISION, MACHINE LEARNING, AND RISK ASSESSMENT FOR COMPREHENSIVE CORROSION EVALUATION"}],"uid":"27707","created_gmt":"2026-03-19 19:19:17","changed_gmt":"2026-03-19 19:19:35","author":"Tatianna Richardson","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2026-04-15T12:00:00-04:00","event_time_end":"2026-04-15T14:00:00-04:00","event_time_end_last":"2026-04-15T14:00:00-04:00","gmt_time_start":"2026-04-15 16:00:00","gmt_time_end":"2026-04-15 18:00:00","gmt_time_end_last":"2026-04-15 18:00:00","rrule":null,"timezone":"America\/New_York"},"location":"Pettit 102A\/ MS Teams Meeting","extras":[],"groups":[{"id":"221981","name":"Graduate Studies"}],"categories":[],"keywords":[{"id":"100811","name":"Phd Defense"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1788","name":"Other\/Miscellaneous"}],"invited_audience":[{"id":"78771","name":"Public"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}