DDR-ID: Dual Deep Reconstruction Networks Based Image Decomposition for Anomaly Detection

TL;DR

DDR-ID introduces a novel dual deep reconstruction network approach that decomposes images into normal and residual parts, improving anomaly detection accuracy by leveraging both latent space and reconstruction errors, validated on multiple datasets.

Contribution

The paper presents a new anomaly detection method using dual deep networks with joint loss optimization for better discrimination of normal and abnormal images.

Findings

Outperforms existing methods on MNIST, CIFAR-10, and Endosome datasets.

Effective in adversarial attack detection on GTSRB dataset.

Utilizes dual decomposition for improved anomaly scoring.

Abstract

One pivot challenge for image anomaly (AD) detection is to learn discriminative information only from normal class training images. Most image reconstruction based AD methods rely on the discriminative capability of reconstruction error. This is heuristic as image reconstruction is unsupervised without incorporating normal-class-specific information. In this paper, we propose an AD method called dual deep reconstruction networks based image decomposition (DDR-ID). The networks are trained by jointly optimizing for three losses: the one-class loss, the latent space constrain loss and the reconstruction loss. After training, DDR-ID can decompose an unseen image into its normal class and the residual components, respectively. Two anomaly scores are calculated to quantify the anomalous degree of the image in either normal class latent space or reconstruction image space. Thereby, anomaly detection can be performed via thresholding the anomaly score. The experiments demonstrate that DDR-ID outperforms multiple related benchmarking methods in image anomaly detection using MNIST, CIFAR-10 and Endosome datasets and adversarial attack detection using GTSRB dataset.