A Multi-Scale A Contrario method for Unsupervised Image Anomaly Detection

Matias Tailanian, Pablo Muse, Alvaro Pardo

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

5 Citas (Scopus)

Resumen

Anomalies can be defined as any non-random structure which deviates from normality. Anomaly detection methods reported in the literature are numerous and diverse, as what is considered anomalous usually varies depending on particular scenarios and applications. In this work we propose an a contrario framework to detect anomalies in images applying statistical analysis to feature maps obtained via convolutions. We evaluate filters learned from the image under analysis via patch PCA and the feature maps obtained from a pre-trained deep neural network (Resnet). The proposed method is multi-scale and fully unsupervised, and is able to detect anomalies in a wide variety of scenarios. While the end goal of this work is the detection of subtle defects in leather samples for the automotive industry, we show that the same algorithm achieves state-of-the-art results in public anomalies datasets.

Idioma originalInglés
Título de la publicación alojadaProceedings - 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021
EditoresM. Arif Wani, Ishwar K. Sethi, Weisong Shi, Guangzhi Qu, Daniela Stan Raicu, Ruoming Jin
EditorialInstitute of Electrical and Electronics Engineers Inc.
Páginas179-184
Número de páginas6
ISBN (versión digital)9781665443371
DOI
EstadoPublicada - 2021
Evento20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021 - Virtual, Online
Duración: 13 dic. 202116 dic. 2021

Serie de la publicación

NombreProceedings - 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021

Conferencia

Conferencia20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021
País/TerritorioUnited States
CiudadVirtual, Online
Período13/12/2116/12/21

Huella

Profundice en los temas de investigación de 'A Multi-Scale A Contrario method for Unsupervised Image Anomaly Detection'. En conjunto forman una huella única.

Citar esto