{"id":{"repo_id":"cau-kiel","oai_identifier":"oai:macau.uni-kiel.de:macau_mods_00008059"},"canonical_url":"https://search.dev.ndltd.org/etd/cau-kiel/oai:macau.uni-kiel.de:macau_mods_00008059","repository":{"repo_id":"cau-kiel","name":"Christian-Albrechts Universität Kiel","base_url":"https://macau.uni-kiel.de/servlets/OAIDataProvider"},"display":{"title":"Underwater Refractive Camera Calibration and 3D Scene Reconstruction","abstract":"Underwater vision systems play a critical role in advancing oceanic applications, including quantifying marine parameters, spatially mapping the seafloor, and digitally documenting plankton. Given that the majority of Earth's surface lies in deep-sea regions, achieving these objectives necessitates the use of unmanned vehicles equipped with waterproof camera-light systems. Despite prior research addressing related challenges, these tasks remain challenging due to two key issues: (1) geometric distortions caused by distance-dependent light refraction when the light rays travel through the housing interfaces in a non-orthogonal manner. (2) radiometric degradation due to light attenuation and scattering in the underwater medium, and inconsistency introduced by camera-attached illumination setups. The first part of the work focuses on refractive camera calibration and 3D reconstruction. It starts with investigating geometric models and calibration methods for dome-port camera systems, then introduces an open-source refractive Structure-from-Motion (SfM) solution that integrates refraction throughout the reconstruction pipeline. Finally, a comprehensive offline monocular visual mapping system is established to enable drift-free, efficient 3D reconstruction of large-area of seafloor from datasets collected by deep-sea robots. The second part addresses radiometric restoration, developing approaches to model the camera-attached illumination source in conjunction with the underwater image formation model for scene color recovery. Finally, this research demonstrates novel oceanic applications enabled by 3D visual computing techniques, including quantifying gas release from seafloor vents and digitally documenting marine plankton for ecological monitoring.","abstract_html":"Underwater vision systems play a critical role in advancing oceanic applications, including quantifying marine parameters, spatially mapping the seafloor, and digitally documenting plankton. Given that the majority of Earth's surface lies in deep-sea regions, achieving these objectives necessitates the use of unmanned vehicles equipped with waterproof camera-light systems. Despite prior research addressing related challenges, these tasks remain challenging due to two key issues: (1) geometric distortions caused by distance-dependent light refraction when the light rays travel through the housing interfaces in a non-orthogonal manner. (2) radiometric degradation due to light attenuation and scattering in the underwater medium, and inconsistency introduced by camera-attached illumination setups. The first part of the work focuses on refractive camera calibration and 3D reconstruction. It starts with investigating geometric models and calibration methods for dome-port camera systems, then introduces an open-source refractive Structure-from-Motion (SfM) solution that integrates refraction throughout the reconstruction pipeline. Finally, a comprehensive offline monocular visual mapping system is established to enable drift-free, efficient 3D reconstruction of large-area of seafloor from datasets collected by deep-sea robots. The second part addresses radiometric restoration, developing approaches to model the camera-attached illumination source in conjunction with the underwater image formation model for scene color recovery. Finally, this research demonstrates novel oceanic applications enabled by 3D visual computing techniques, including quantifying gas release from seafloor vents and digitally documenting marine plankton for ecological monitoring.","abstract_has_math":false,"creators":["She, Mengkun"],"institution":"Christian-Albrechts-Universität zu Kiel","degree_name":null,"degree_level":"thesis.doctoral","degree_discipline":null,"degree_department":null,"school":null,"contributors":["Köser, Kevin","Birk, Andreas","Mester, Rudolf"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026-02-27","date_published":"2026-02-27","updated_at":"2026-07-24T01:35:31Z","subjects":["Underwater Computer Vision","Refractive Camera Calibration","Refractive 3D reconstruction","Underwater Image Restoration","Underwater Illumination Learning"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://macau.uni-kiel.de/receive/macau_mods_00008059","outbound_label":"Repository record","outbound_source":"source_url"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Köser, Kevin","Birk, Andreas","Mester, Rudolf"]},{"key":"dc:creator","label":"Author","values":["She, Mengkun"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:publisher","label":"Institution","values":["Universitätsbibliothek Kiel"]},{"key":"dc:type","label":"Dc Type","values":["PhDThesis"]},{"key":"thesis:degree_level","label":"Degree Level","values":["thesis.doctoral"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Christian-Albrechts-Universität zu Kiel"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Underwater Computer Vision","Refractive Camera Calibration","Refractive 3D reconstruction","Underwater Image Restoration","Underwater Illumination Learning"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Underwater vision systems play a critical role in advancing oceanic applications, including quantifying marine parameters, spatially mapping the seafloor, and digitally documenting plankton. Given that the majority of Earth's surface lies in deep-sea regions, achieving these objectives necessitates the use of unmanned vehicles equipped with waterproof camera-light systems. Despite prior research addressing related challenges, these tasks remain challenging due to two key issues: (1) geometric distortions caused by distance-dependent light refraction when the light rays travel through the housing interfaces in a non-orthogonal manner. (2) radiometric degradation due to light attenuation and scattering in the underwater medium, and inconsistency introduced by camera-attached illumination setups. The first part of the work focuses on refractive camera calibration and 3D reconstruction. It starts with investigating geometric models and calibration methods for dome-port camera systems, then introduces an open-source refractive Structure-from-Motion (SfM) solution that integrates refraction throughout the reconstruction pipeline. Finally, a comprehensive offline monocular visual mapping system is established to enable drift-free, efficient 3D reconstruction of large-area of seafloor from datasets collected by deep-sea robots. The second part addresses radiometric restoration, developing approaches to model the camera-attached illumination source in conjunction with the underwater image formation model for scene color recovery. Finally, this research demonstrates novel oceanic applications enabled by 3D visual computing techniques, including quantifying gas release from seafloor vents and digitally documenting marine plankton for ecological monitoring."]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Underwater Refractive Camera Calibration and 3D Scene Reconstruction"]}]}],"canonical_facts":{"dc:contributor":["Köser, Kevin","Birk, Andreas","Mester, Rudolf"],"dc:creator":["She, Mengkun"],"dc:description.abstract":["Underwater vision systems play a critical role in advancing oceanic applications, including quantifying marine parameters, spatially mapping the seafloor, and digitally documenting plankton. Given that the majority of Earth's surface lies in deep-sea regions, achieving these objectives necessitates the use of unmanned vehicles equipped with waterproof camera-light systems. Despite prior research addressing related challenges, these tasks remain challenging due to two key issues: (1) geometric distortions caused by distance-dependent light refraction when the light rays travel through the housing interfaces in a non-orthogonal manner. (2) radiometric degradation due to light attenuation and scattering in the underwater medium, and inconsistency introduced by camera-attached illumination setups. The first part of the work focuses on refractive camera calibration and 3D reconstruction. It starts with investigating geometric models and calibration methods for dome-port camera systems, then introduces an open-source refractive Structure-from-Motion (SfM) solution that integrates refraction throughout the reconstruction pipeline. Finally, a comprehensive offline monocular visual mapping system is established to enable drift-free, efficient 3D reconstruction of large-area of seafloor from datasets collected by deep-sea robots. The second part addresses radiometric restoration, developing approaches to model the camera-attached illumination source in conjunction with the underwater image formation model for scene color recovery. Finally, this research demonstrates novel oceanic applications enabled by 3D visual computing techniques, including quantifying gas release from seafloor vents and digitally documenting marine plankton for ecological monitoring."],"dc:format.medium":["application/pdf"],"dc:publisher":["Universitätsbibliothek Kiel"],"dc:subject":["Underwater Computer Vision","Refractive Camera Calibration","Refractive 3D reconstruction","Underwater Image Restoration","Underwater Illumination Learning"],"dc:title":["Underwater Refractive Camera Calibration and 3D Scene Reconstruction"],"dc:type":["PhDThesis"],"thesis:degree_level":["thesis.doctoral"],"thesis:institution_name":["Christian-Albrechts-Universität zu Kiel"]},"updated_at":"2026-07-24T01:35:31Z"}