Industrial product anomaly detection method and system based on generative adversarial network

The invention discloses an industrial product anomaly detection method and system based on a generative adversarial network, and belongs to the technical field of artificial intelligence. Firstly, a generator and a multi-scale-based discriminator are designed, the discriminator aims at correctly dis...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: JIAO AIMING, DU YUZHUO, SHU XINGJIE, ZOU GUANGREN, HE KAI, WANG ZEGUANG, SUN FENGWEI, WANG YANSHENG
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses an industrial product anomaly detection method and system based on a generative adversarial network, and belongs to the technical field of artificial intelligence. Firstly, a generator and a multi-scale-based discriminator are designed, the discriminator aims at correctly discriminating a real product and a generated product as much as possible, and the generator aims at generating an image close to a real industrial product as much as possible. Secondly, designing a bioconfrontation network structure; and finally, performing an experiment on the data set, analyzing and comparing, and then verifying the validity of the method in an abnormal detection task oriented to an industrial scene. A Transform model is adopted to construct a generator and a discriminator based on an anomaly detection method, global context information is better captured, the calculation amount and the local feature capturing capacity are optimized, training data do not need to be labeled, the method is very suita