{"id":{"repo_id":"eiu","oai_identifier":"oai:thekeep.eiu.edu:theses-1702"},"canonical_url":"https://search.ndltd.org/etd/eiu/oai:thekeep.eiu.edu:theses-1702","repository":{"repo_id":"eiu","name":"Eastern Illinois University","base_url":"https://thekeep.eiu.edu/do/oai/"},"display":{"title":"Facial Recognition System Screening Evaluation Methodology For Complexion Biases","abstract":"Over the years, facial recognition systems (FRS) have come under scrutiny from watchdog groups who voice their complaints concerning the potential existence of FRS biases towards certain cultures of people when deployed in security screening situations. Psychology has used the term \"other-race effect\" to describe this occurrence in humans. Researchers of FRS have used the theory behind the other-race effect to describe the occurrence of bias associated with algorithms tested during the Face Recognition Vendor Test of 2006. In this thesis, we develop a scientific methodology based on the factors of illumination, distance, and angle to evaluate whether or not a chosen system exhibits a significant bias towards a light complexion 3D facial model over a medium complexion 3D facial model. In our particular test, chosen parameters, and statistical software (DOE PRO), our methodology shows that there is no statistically significant bias between the sample mean of a light complexion 3D facial model and the sample mean of a medium complexion 3D facial model. Identical statistical results can n9t be guaranteed for all FRS, although the methodology can be applied to any FRS to determine if a complexion bias exists.","abstract_html":"Over the years, facial recognition systems (FRS) have come under scrutiny from watchdog groups who voice their complaints concerning the potential existence of FRS biases towards certain cultures of people when deployed in security screening situations. Psychology has used the term &quot;other-race effect&quot; to describe this occurrence in humans. Researchers of FRS have used the theory behind the other-race effect to describe the occurrence of bias associated with algorithms tested during the Face Recognition Vendor Test of 2006. In this thesis, we develop a scientific methodology based on the factors of illumination, distance, and angle to evaluate whether or not a chosen system exhibits a significant bias towards a light complexion 3D facial model over a medium complexion 3D facial model. In our particular test, chosen parameters, and statistical software (DOE PRO), our methodology shows that there is no statistically significant bias between the sample mean of a light complexion 3D facial model and the sample mean of a medium complexion 3D facial model. Identical statistical results can n9t be guaranteed for all FRS, although the methodology can be applied to any FRS to determine if a complexion bias exists.","abstract_has_math":false,"creators":["Harris, Harold"],"institution":null,"degree_name":"Master of Science (MS)","degree_level":null,"degree_discipline":"School of Technology","degree_department":null,"school":null,"contributors":["Rigoberto Chinchilla","Dr. Izadi","Dr. McDonald"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T07:54:45Z","subjects":["Science and Technology Studies"],"languages":[],"rights":[],"rights_urls":["http://creativecommons.org/licenses/by-nc-nd/3.0/"],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://thekeep.eiu.edu/theses/703","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://thekeep.eiu.edu/theses/703","record_label":"Repository record","record_kind":"record","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://thekeep.eiu.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Athekeep.eiu.edu%3Atheses-1702","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Rigoberto Chinchilla","Dr. Izadi","Dr. McDonald"]},{"key":"dc:creator","label":"Author","values":["Harris, Harold"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["1970-01-01T08:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2010-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["School of Technology"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science (MS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Science and Technology Studies"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights.license","label":"Dc Rights License","values":["http://creativecommons.org/licenses/by-nc-nd/3.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://thekeep.eiu.edu/theses/703"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Over the years, facial recognition systems (FRS) have come under scrutiny from watchdog groups who voice their complaints concerning the potential existence of FRS biases towards certain cultures of people when deployed in security screening situations. Psychology has used the term \"other-race effect\" to describe this occurrence in humans. Researchers of FRS have used the theory behind the other-race effect to describe the occurrence of bias associated with algorithms tested during the Face Recognition Vendor Test of 2006. In this thesis, we develop a scientific methodology based on the factors of illumination, distance, and angle to evaluate whether or not a chosen system exhibits a significant bias towards a light complexion 3D facial model over a medium complexion 3D facial model. In our particular test, chosen parameters, and statistical software (DOE PRO), our methodology shows that there is no statistically significant bias between the sample mean of a light complexion 3D facial model and the sample mean of a medium complexion 3D facial model. Identical statistical results can n9t be guaranteed for all FRS, although the methodology can be applied to any FRS to determine if a complexion bias exists."]},{"key":"dc:title","label":"Title","values":["Facial Recognition System Screening Evaluation Methodology For Complexion Biases"]}]}],"canonical_facts":{"dc:contributor":["Rigoberto Chinchilla","Dr. Izadi","Dr. McDonald"],"dc:creator":["Harris, Harold"],"dc:date.available":["1970-01-01T08:00:00Z"],"dc:date.created":["2010-01-01T08:00:00Z"],"dc:description.abstract":["Over the years, facial recognition systems (FRS) have come under scrutiny from watchdog groups who voice their complaints concerning the potential existence of FRS biases towards certain cultures of people when deployed in security screening situations. Psychology has used the term \"other-race effect\" to describe this occurrence in humans. Researchers of FRS have used the theory behind the other-race effect to describe the occurrence of bias associated with algorithms tested during the Face Recognition Vendor Test of 2006. In this thesis, we develop a scientific methodology based on the factors of illumination, distance, and angle to evaluate whether or not a chosen system exhibits a significant bias towards a light complexion 3D facial model over a medium complexion 3D facial model. In our particular test, chosen parameters, and statistical software (DOE PRO), our methodology shows that there is no statistically significant bias between the sample mean of a light complexion 3D facial model and the sample mean of a medium complexion 3D facial model. Identical statistical results can n9t be guaranteed for all FRS, although the methodology can be applied to any FRS to determine if a complexion bias exists."],"dc:identifier":["https://thekeep.eiu.edu/theses/703"],"dc:rights.license":["http://creativecommons.org/licenses/by-nc-nd/3.0/"],"dc:subject":["Science and Technology Studies"],"dc:title":["Facial Recognition System Screening Evaluation Methodology For Complexion Biases"],"thesis:degree_discipline":["School of Technology"],"thesis:degree_name":["Master of Science (MS)"]},"updated_at":"2026-11-19T07:54:45Z"}