Industrial anomaly detection with normalizing flows / Marco Rudolph

Von:

Rudolph, Marco, 1996- [VerfasserIn]

Resource type: Ressourcentyp: Buch (Online)Buch (Online)Sprache: Englisch Reihen: Fortschritt-Berichte VDI. Reihe 10Informatik/Kommunikation ; Nr. 886Verlag: Düsseldorf : VDI Verlag GmbH, [2024]Copyright-Datum: © 2024Beschreibung: 1 Online-Ressource (XIV, 126 Seiten) : IllustrationenISBN:

9783186886101

Genre/Form:

Hochschulschrift

Andere physische Formen: 9783183886104 | Erscheint auch als: 978-3-18-388610-4 Druck-AusgabeDOI: DOI: 10.51202/9783186886101Online-Ressourcen:

Zugang im Netz des KIT

Hochschulschriftenvermerk: Dissertation - Gottfried Wilhelm Leibniz Universität Hannover, 2024 Zusammenfassung: This thesis addresses deep learning-based methods for automatic anomaly detection in an industrial context. It involves image- or sensor-based detection of defects in the production process that can affect the quality of products. Automating this task provides a reliable and cost-effective alternative to humans, who perform this task manually by sighting. Since this setup has special requirements such as detecting previously unknown defects that traditional approaches cannot fulfill, this paper presents anomaly detection methods that learn without any examples of anomalies and include only normal data in the training process. Most of our proposed methods address the problem from a statistical perspective. Based on a deep-learning-based density estimation of the normal data, it is assumed that anomalies are considered unlikely according to the modeled distribution. The density estimation is performed by socalled Normalizing Flows, which, in contrast to conventional neural networks, can model a formally valid probability distribution due to their bijective mapping.PPN: PPN: 189468821XPackage identifier: Produktsigel: ZDB-281-VDI | ZDB-281-VDG

Exemplare ( 0 )

Dieser Titel hat keine Exemplare