English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In the anomaly detection field, the scarcity of anomalous samples hasdirected the current research emphasis towards unsupervised anomaly detection.While these unsupervised anomaly detection methods offer convenience, they alsooverlook the crucial prior information embedded within anomalous samples.Moreover, among numerous deep learning methods, supervised methods generallyexhibit superior performance compared to unsupervised methods. Considering thereasons mentioned above, we propose a self-supervised anomaly detectionapproach that combines contrastive learning with 2D-Flow to achieve moreprecise detection outcomes and expedited inference processes. On one hand, weintroduce a novel approach to anomaly synthesis, yielding anomalous samples inaccordance with authentic industrial scenarios, alongside their surrogateannotations. On the other hand, having obtained a substantial number ofanomalous samples, we enhance the 2D-Flow framework by incorporatingcontrastive learning, leveraging diverse proxy tasks to fine-tune the network.Our approach enables the network to learn more precise mapping relationshipsfrom self-generated labels while retaining the lightweight characteristics ofthe 2D-Flow. Compared to mainstream unsupervised approaches, ourself-supervised method demonstrates superior detection accuracy, feweradditional model parameters, and faster inference speed. Furthermore, theentire training and inference process is end-to-end. Our approach showcases newstate-of-the-art results, achieving a performance of 99.6\% in image-levelAUROC on the MVTecAD dataset and 96.8\% in image-level AUROC on the BTADdataset.

CL-Flow:Strengthening the Normalizing Flows by Contrastive Learning for Better Anomaly Detection