GRD-Net: Generative-Reconstructive-Discriminative Anomaly Detection with Region of Interest Attention Module

TL;DR

GRD-Net is a novel anomaly detection architecture combining generative, reconstructive, and discriminative components with ROI attention, improving defect localization in industrial visual inspection tasks.

Contribution

The paper introduces a new GAN-based architecture with ROI attention for anomaly detection, reducing reliance on traditional post-processing methods and enhancing localization accuracy.

Findings

Effective on MVTec datasets with high accuracy.

Successfully applied to pharmaceutical BFS vial images.

Reduces need for pre-processing algorithms.

Abstract

Anomaly detection is nowadays increasingly used in industrial applications and processes. One of the main fields of the appliance is the visual inspection for surface anomaly detection, which aims to spot regions that deviate from regularity and consequently identify abnormal products. Defect localization is a key task, that usually is achieved using a basic comparison between generated image and the original one, implementing some blob-analysis or image-editing algorithms, in the post-processing step, which is very biased towards the source dataset, and they are unable to generalize. Furthermore, in industrial applications, the totality of the image is not always interesting but could be one or some regions of interest (ROIs), where only in those areas there are relevant anomalies to be spotted. For these reasons, we propose a new architecture composed by two blocks. The first block is a Generative Adversarial Network (GAN), based on a residual autoencoder (ResAE), to perform reconstruction and denoising processes, while the second block produces image segmentation, spotting defects. This method learns from a dataset composed of good products and generated synthetic defects. The discriminative network is trained using a ROI for each image contained in the training dataset. The network will learn in which area anomalies are relevant. This approach guarantees the reduction of using pre-processing algorithms, formerly developed with blob-analysis and image-editing procedures. To test our model we used challenging MVTec anomaly detection datasets and an industrial large dataset of pharmaceutical BFS strips of vials. This set constitutes a more realistic use case of the aforementioned network.