Anomaly Detection in Image Datasets Using Convolutional Neural Networks, Center Loss, and Mahalanobis Distance

TL;DR

This paper introduces methods for detecting out-of-distribution images in datasets using CNNs with center loss and Mahalanobis distance, enabling simultaneous classification and anomaly detection.

Contribution

It extends neural networks with center loss to improve out-of-distribution detection in image datasets, combining classification and anomaly detection in a unified approach.

Findings

Effective detection of out-of-distribution samples on MNIST and ImageNet-30.

Improved anomaly detection performance with center loss integration.

Analysis of deep feature distributions for anomaly detection.

Abstract

User activities generate a significant number of poor-quality or irrelevant images and data vectors that cannot be processed in the main data processing pipeline or included in the training dataset. Such samples can be found with manual analysis by an expert or with anomalous detection algorithms. There are several formal definitions for the anomaly samples. For neural networks, the anomalous is usually defined as out-of-distribution samples. This work proposes methods for supervised and semi-supervised detection of out-of-distribution samples in image datasets. Our approach extends a typical neural network that solves the image classification problem. Thus, one neural network after extension can solve image classification and anomalous detection problems simultaneously. Proposed methods are based on the center loss and its effect on a deep feature distribution in a last hidden layer of the neural network. This paper provides an analysis of the proposed methods for the LeNet and EfficientNet-B0 on the MNIST and ImageNet-30 datasets.