{"id":{"repo_id":"wvu","oai_identifier":"oai:researchrepository.wvu.edu:etd-1155"},"canonical_url":"https://search.dev.ndltd.org/etd/wvu/oai:researchrepository.wvu.edu:etd-1155","repository":{"repo_id":"wvu","name":"West Virginia University","base_url":"https://researchrepository.wvu.edu/do/oai/"},"display":{"title":"A statistical approach for shadow detection using spatio-temporal contexts","abstract":"Background subtraction is an important step used to segment moving regions in surveillance videos. However, cast shadows are often falsely labeled as foreground objects, which may severely degrade the accuracy of object localization and detection. Effective shadow detection is necessary for accurate foreground segmentation, especially for outdoor scenes. Based on the characteristics of shadows, such as luminance reduction, chromaticity consistency and texture consistency, we introduce a nonparametric framework for modeling surface behavior under cast shadows. To each pixel, we assign a potential shadow value with a confidence weight, indicating the probability that the pixel location is an actual shadow point. Given an observed RGB value for a pixel in a new frame, we use its recent spatio-temporal context to compute an expected shadow RGB value. The similarity between the observed and the expected shadow RGB values determines whether a pixel position is a true shadow. Experimental results show the performance of the proposed method on a suite of standard indoor and outdoor video sequences.","abstract_html":"Background subtraction is an important step used to segment moving regions in surveillance videos. However, cast shadows are often falsely labeled as foreground objects, which may severely degrade the accuracy of object localization and detection. Effective shadow detection is necessary for accurate foreground segmentation, especially for outdoor scenes. Based on the characteristics of shadows, such as luminance reduction, chromaticity consistency and texture consistency, we introduce a nonparametric framework for modeling surface behavior under cast shadows. To each pixel, we assign a potential shadow value with a confidence weight, indicating the probability that the pixel location is an actual shadow point. Given an observed RGB value for a pixel in a new frame, we use its recent spatio-temporal context to compute an expected shadow RGB value. The similarity between the observed and the expected shadow RGB values determines whether a pixel position is a true shadow. Experimental results show the performance of the proposed method on a suite of standard indoor and outdoor video sequences.","abstract_has_math":false,"creators":["Liu, Yiyang"],"institution":null,"degree_name":"MS","degree_level":"Thesis","degree_discipline":"Lane Department of Computer Science and Electrical Engineering","degree_department":null,"school":null,"contributors":["Donald Adjeroh","Xin Li","Arun Ross."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2011,"date_issued":"2011-12-01T08:00:00Z","date_published":"2011-12-01T08:00:00Z","updated_at":"2026-07-24T06:14:02Z","subjects":["Computer science"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://researchrepository.wvu.edu/etd/152"],"render_values":[{"text":"https://researchrepository.wvu.edu/etd/152","href":"https://researchrepository.wvu.edu/etd/152","code":true}]}]},"links":{"outbound_url":"https://doi.org/10.33915/etd.152","outbound_label":"DOI","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Donald Adjeroh","Xin Li","Arun Ross."]},{"key":"dc:creator","label":"Author","values":["Liu, Yiyang"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2018-10-29T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Lane Department of Computer Science and Electrical Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["MS"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Computer science"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://doi.org/10.33915/etd.152","https://researchrepository.wvu.edu/etd/152"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Background subtraction is an important step used to segment moving regions in surveillance videos. However, cast shadows are often falsely labeled as foreground objects, which may severely degrade the accuracy of object localization and detection. Effective shadow detection is necessary for accurate foreground segmentation, especially for outdoor scenes. Based on the characteristics of shadows, such as luminance reduction, chromaticity consistency and texture consistency, we introduce a nonparametric framework for modeling surface behavior under cast shadows. To each pixel, we assign a potential shadow value with a confidence weight, indicating the probability that the pixel location is an actual shadow point. Given an observed RGB value for a pixel in a new frame, we use its recent spatio-temporal context to compute an expected shadow RGB value. The similarity between the observed and the expected shadow RGB values determines whether a pixel position is a true shadow. Experimental results show the performance of the proposed method on a suite of standard indoor and outdoor video sequences."]},{"key":"dc:title","label":"Title","values":["A statistical approach for shadow detection using spatio-temporal contexts"]}]}],"canonical_facts":{"dc:contributor":["Donald Adjeroh","Xin Li","Arun Ross."],"dc:creator":["Liu, Yiyang"],"dc:date.available":["2018-10-29T07:00:00Z"],"dc:description.abstract":["Background subtraction is an important step used to segment moving regions in surveillance videos. However, cast shadows are often falsely labeled as foreground objects, which may severely degrade the accuracy of object localization and detection. Effective shadow detection is necessary for accurate foreground segmentation, especially for outdoor scenes. Based on the characteristics of shadows, such as luminance reduction, chromaticity consistency and texture consistency, we introduce a nonparametric framework for modeling surface behavior under cast shadows. To each pixel, we assign a potential shadow value with a confidence weight, indicating the probability that the pixel location is an actual shadow point. Given an observed RGB value for a pixel in a new frame, we use its recent spatio-temporal context to compute an expected shadow RGB value. The similarity between the observed and the expected shadow RGB values determines whether a pixel position is a true shadow. Experimental results show the performance of the proposed method on a suite of standard indoor and outdoor video sequences."],"dc:identifier":["https://doi.org/10.33915/etd.152","https://researchrepository.wvu.edu/etd/152"],"dc:subject":["Computer science"],"dc:title":["A statistical approach for shadow detection using spatio-temporal contexts"],"thesis:degree_discipline":["Lane Department of Computer Science and Electrical Engineering"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["MS"]},"updated_at":"2026-07-24T06:14:02Z"}