Heuristic Hyperparameter Choice for Image Anomaly Detection

TL;DR

This paper proposes a heuristic method for selecting hyperparameters in Negated Principal Component Analysis to reduce feature dimensions in image anomaly detection, improving efficiency without sacrificing performance.

Contribution

It introduces a novel heuristic for hyperparameter tuning in NPCA, tailored for image anomaly detection to optimize feature reduction.

Findings

Effective dimension reduction with minimal components

Maintains high anomaly detection performance

Reduces computational cost

Abstract

Anomaly detection (AD) in images is a fundamental computer vision problem by deep learning neural network to identify images deviating significantly from normality. The deep features extracted from pretrained models have been proved to be essential for AD based on multivariate Gaussian distribution analysis. However, since models are usually pretrained on a large dataset for classification tasks such as ImageNet, they might produce lots of redundant features for AD, which increases computational cost and degrades the performance. We aim to do the dimension reduction of Negated Principal Component Analysis (NPCA) for these features. So we proposed some heuristic to choose hyperparameter of NPCA algorithm for getting as fewer components of features as possible while ensuring a good performance.