Budde, Lina E. ; Collmar, David ; Sörgel, Uwe ; Iwaszczuk, Dorota
Hrsg.: Kersten, Thomas P. ; Tilly, Nora ; Deutsche Gesellschaft für Photogrammetrie, Fernerkundung und Geoinformation (DGPF) (2024)
Investigating the Relationship between Image Quality
and Crowdsourcing for Labeling.
44. Wissenschaftlich-Technische Jahrestagung der DGPF. Remagen (13.03.2024 -14.03.2024)
doi: 10.24407/KXP:1884384242
Konferenzveröffentlichung, Bibliographie
Kurzbeschreibung (Abstract)
The quality of training data and the performance of machine learning approaches using those data stand in direct correlation: Even the best algorithms are not able to compensate low-quality training data, making the use of high-quality training data essential. However, generating high-quality training data is not trivial, especially as these data are required in large quantities. A common approach is the use of paid crowdsourcing, where the resulting label quality is contingent upon two key factors: the performance of the workers and the intrinsic quality of the initial dataset utilized for these tasks. Limiting factors, such as poor radiometry that leads to low-quality images, must be taken into consideration when evaluating data acquired through crowdsourcing. We examine the relationship between image quality and the output performance of crowdsourcing tasks in the context of tree crown detection.
Typ des Eintrags: | Konferenzveröffentlichung |
---|---|
Erschienen: | 2024 |
Herausgeber: | Kersten, Thomas P. ; Tilly, Nora |
Autor(en): | Budde, Lina E. ; Collmar, David ; Sörgel, Uwe ; Iwaszczuk, Dorota |
Art des Eintrags: | Bibliographie |
Titel: | Investigating the Relationship between Image Quality and Crowdsourcing for Labeling |
Sprache: | Englisch |
Publikationsjahr: | März 2024 |
Ort: | Stuttgart |
Verlag: | Geschäftsstelle der DGPF |
Buchtitel: | DGPF-Jahrestagung 2024: Stadt, Land, Fluss - Daten vernetzen : Beiträge |
Reihe: | Publikationen der Deutschen Gesellschaft für Photogrammetrie, Fernerkundung und Geoinformation e.V. |
Band einer Reihe: | 32 |
Veranstaltungstitel: | 44. Wissenschaftlich-Technische Jahrestagung der DGPF |
Veranstaltungsort: | Remagen |
Veranstaltungsdatum: | 13.03.2024 -14.03.2024 |
DOI: | 10.24407/KXP:1884384242 |
Zugehörige Links: | |
Kurzbeschreibung (Abstract): | The quality of training data and the performance of machine learning approaches using those data stand in direct correlation: Even the best algorithms are not able to compensate low-quality training data, making the use of high-quality training data essential. However, generating high-quality training data is not trivial, especially as these data are required in large quantities. A common approach is the use of paid crowdsourcing, where the resulting label quality is contingent upon two key factors: the performance of the workers and the intrinsic quality of the initial dataset utilized for these tasks. Limiting factors, such as poor radiometry that leads to low-quality images, must be taken into consideration when evaluating data acquired through crowdsourcing. We examine the relationship between image quality and the output performance of crowdsourcing tasks in the context of tree crown detection. |
Fachbereich(e)/-gebiet(e): | 13 Fachbereich Bau- und Umweltingenieurwissenschaften 13 Fachbereich Bau- und Umweltingenieurwissenschaften > Institut für Geodäsie 13 Fachbereich Bau- und Umweltingenieurwissenschaften > Institut für Geodäsie > Fernerkundung und Bildanalyse |
Hinterlegungsdatum: | 13 Jun 2024 05:57 |
Letzte Änderung: | 01 Jul 2024 09:05 |
PPN: | 519476638 |
Export: | |
Suche nach Titel in: | TUfind oder in Google |
![]() |
Frage zum Eintrag |
Optionen (nur für Redakteure)
![]() |
Redaktionelle Details anzeigen |