{"id":{"repo_id":"mtu","oai_identifier":"oai:digitalcommons.mtu.edu:etdr-3055"},"canonical_url":"https://search.ndltd.org/etd/mtu/oai:digitalcommons.mtu.edu:etdr-3055","repository":{"repo_id":"mtu","name":"Michigan Technological University","base_url":"https://digitalcommons.mtu.edu/do/oai/"},"display":{"title":"EVALUATING CHALLENGES AND SOLUTIONS FOR BUCKTHORN CLASSIFICATION IN SHADOWED ENVIRONMENTS","abstract":"Invasive species, such as buckthorns, pose significant ecological threats by displacing native vegetation and reducing biodiversity. This study examines the impact of shadows on the classification accuracy of buckthorns using drone-based multispectral imagery collected in a forested area near Michigan Technological University. Shadows impacted approximately 70% of the imagery, notably distorting reflectance in key spectral bands such as near-infrared (NIR) and red edge. The study evaluated vegetation indices like NDVI and the performance of machine learning models, specifically the Random Forest classifier, under these shadowed conditions. Conventional shadow correction techniques, including histogram normalization, Shadow Index (SI), and Inverse Distance Weighting (IDW) interpolation, provided only marginal improvements. The highest classification accuracy achieved was 49.5%, with a Kappa coefficient of 0.24. These findings highlight the challenges of utilizing single-date multispectral imagery in heavily shadowed environments. The study recommends exploring advanced techniques such as Hue Saturation Value (HSV) correction, multi-temporal data fusion, and deep learning approaches to enhance vegetation classification.","abstract_html":"Invasive species, such as buckthorns, pose significant ecological threats by displacing native vegetation and reducing biodiversity. This study examines the impact of shadows on the classification accuracy of buckthorns using drone-based multispectral imagery collected in a forested area near Michigan Technological University. Shadows impacted approximately 70% of the imagery, notably distorting reflectance in key spectral bands such as near-infrared (NIR) and red edge. The study evaluated vegetation indices like NDVI and the performance of machine learning models, specifically the Random Forest classifier, under these shadowed conditions. Conventional shadow correction techniques, including histogram normalization, Shadow Index (SI), and Inverse Distance Weighting (IDW) interpolation, provided only marginal improvements. The highest classification accuracy achieved was 49.5%, with a Kappa coefficient of 0.24. These findings highlight the challenges of utilizing single-date multispectral imagery in heavily shadowed environments. The study recommends exploring advanced techniques such as Hue Saturation Value (HSV) correction, multi-temporal data fusion, and deep learning approaches to enhance vegetation classification.","abstract_has_math":false,"creators":["Tangwam, Moses"],"institution":null,"degree_name":"Master of Geographic Information Science","degree_level":"Open Access Master's Report","degree_discipline":"College of Forest Resources and Environmental Science","degree_department":null,"school":null,"contributors":["Mickey P. Jarvi","Parth Parimalbhai Bhatt"],"advisors":[],"committee_chairs":[],"committee_members":["Michael D. Hyslop","Sigred Resh"],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:19:12Z","subjects":[],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.mtu.edu/etdr/1937"],"render_values":[{"text":"https://digitalcommons.mtu.edu/etdr/1937","href":"https://digitalcommons.mtu.edu/etdr/1937","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/10.37099/mtu.dc.etdr/1937","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/10.37099/mtu.dc.etdr/1937","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%3Aetdr-3055","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Mickey P. Jarvi","Parth Parimalbhai Bhatt"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Michael D. Hyslop","Sigred Resh"]},{"key":"dc:creator","label":"Author","values":["Tangwam, Moses"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.created","label":"Dc Date Created","values":["2025-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["College of Forest Resources and Environmental Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Open Access Master's Report"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Geographic Information Science"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["10.37099/mtu.dc.etdr/1937","https://digitalcommons.mtu.edu/etdr/1937"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Invasive species, such as buckthorns, pose significant ecological threats by displacing native vegetation and reducing biodiversity. This study examines the impact of shadows on the classification accuracy of buckthorns using drone-based multispectral imagery collected in a forested area near Michigan Technological University. Shadows impacted approximately 70% of the imagery, notably distorting reflectance in key spectral bands such as near-infrared (NIR) and red edge. The study evaluated vegetation indices like NDVI and the performance of machine learning models, specifically the Random Forest classifier, under these shadowed conditions. Conventional shadow correction techniques, including histogram normalization, Shadow Index (SI), and Inverse Distance Weighting (IDW) interpolation, provided only marginal improvements. The highest classification accuracy achieved was 49.5%, with a Kappa coefficient of 0.24. These findings highlight the challenges of utilizing single-date multispectral imagery in heavily shadowed environments. The study recommends exploring advanced techniques such as Hue Saturation Value (HSV) correction, multi-temporal data fusion, and deep learning approaches to enhance vegetation classification."]},{"key":"dc:title","label":"Title","values":["EVALUATING CHALLENGES AND SOLUTIONS FOR BUCKTHORN CLASSIFICATION IN SHADOWED ENVIRONMENTS"]}]}],"canonical_facts":{"dc:contributor":["Mickey P. Jarvi","Parth Parimalbhai Bhatt"],"dc:contributor.committeemember":["Michael D. Hyslop","Sigred Resh"],"dc:creator":["Tangwam, Moses"],"dc:date.created":["2025-01-01T08:00:00Z"],"dc:description.abstract":["Invasive species, such as buckthorns, pose significant ecological threats by displacing native vegetation and reducing biodiversity. This study examines the impact of shadows on the classification accuracy of buckthorns using drone-based multispectral imagery collected in a forested area near Michigan Technological University. Shadows impacted approximately 70% of the imagery, notably distorting reflectance in key spectral bands such as near-infrared (NIR) and red edge. The study evaluated vegetation indices like NDVI and the performance of machine learning models, specifically the Random Forest classifier, under these shadowed conditions. Conventional shadow correction techniques, including histogram normalization, Shadow Index (SI), and Inverse Distance Weighting (IDW) interpolation, provided only marginal improvements. The highest classification accuracy achieved was 49.5%, with a Kappa coefficient of 0.24. These findings highlight the challenges of utilizing single-date multispectral imagery in heavily shadowed environments. The study recommends exploring advanced techniques such as Hue Saturation Value (HSV) correction, multi-temporal data fusion, and deep learning approaches to enhance vegetation classification."],"dc:identifier":["10.37099/mtu.dc.etdr/1937","https://digitalcommons.mtu.edu/etdr/1937"],"dc:title":["EVALUATING CHALLENGES AND SOLUTIONS FOR BUCKTHORN CLASSIFICATION IN SHADOWED ENVIRONMENTS"],"thesis:degree_discipline":["College of Forest Resources and Environmental Science"],"thesis:degree_level":["Open Access Master's Report"],"thesis:degree_name":["Master of Geographic Information Science"]},"updated_at":"2026-11-19T08:19:12Z"}