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

Anomaly detection (AD) in images, identifying significant deviations fromnormality, is a critical issue in computer vision. This paper introduces anovel approach to dimensionality reduction for AD using pre-trainedconvolutional neural network (CNN) that incorporate EfficientNet models. Weinvestigate the importance of component selection and propose two types of treesearch approaches, both employing a greedy strategy, for optimal eigencomponentselection. Our study conducts three main experiments to evaluate theeffectiveness of our approach. The first experiment explores the influence oftest set performance on component choice, the second experiment examines theperformance when we train on one anomaly type and evaluate on all other types,and the third experiment investigates the impact of using a minimum number ofimages for training and selecting them based on anomaly types. Our approachaims to find the optimal subset of components that deliver the highestperformance score, instead of focusing solely on the proportion of varianceexplained by each component and also understand the components behaviour indifferent settings. Our results indicate that the proposed method surpassesboth Principal Component Analysis (PCA) and Negated Principal ComponentAnalysis (NPCA) in terms of detection accuracy, even when using fewercomponents. Thus, our approach provides a promising alternative to conventionaldimensionality reduction techniques in AD, and holds potential to enhance theefficiency and effectiveness of AD systems.

Gaussian Image Anomaly Detection with Greedy Eigencomponent Selection