Abstract:Anomaly detection is a crucial means to detect unbalanced and multiclass abnormal data in industrial products. The existing anomaly detection model is usually suitable for products with a simple or regular background texture. However, there is a lack of a universal surface defect anomaly detection framework for railway products when facing the challenges: the complex background texture, huge intraclass variation, uneven distribution of interclass, and weak texture-featured defects. To fill this gap, an innovative generative adversarial network based on adaptive pyramid graph (APG) and variation residuals (APGVR-GAN) is proposed, aiming to improve the robustness of anomaly detection in railway products and other complex industrial supplies. First, the APG module is embedded in the encoder–decoder–encoder pattern, capturing the correlation description between neighbor regions, which is utilized to enhance the detection of abnormal defects with weak texture. Next, the variation residual module is employed to enhance the expression of various normal samples in the latent space and improve the identification ability for abnormal samples. Then, the dual-probability prototype loss is proposed to make different normal samples have more concentrated expression and more similar probability distribution centers in latent space. Finally, an adaptive focal-gate loss and a regularized log-likelihood loss are designed to overcome the imbalance problem in training samples with different background information. The effectiveness of the model is verified on three new railway datasets and three other industrial public datasets. The results show that in anomaly detection, APGVR-GAN on the three railway datasets, Northeastern University (NEU)-the rail surface defects dataset (RSDD)-1 dataset, NEU-RSDD-1 dataset, and RSDD area under curve (AUC) [that is the receiver operating characteristic curve (ROC)], can reach 0.89, 0.81, and 0.87, and its detection accuracy (ACC) reached 0.97, 0.96, and 0.92, which is much higher than other comparative anomaly detection algorithms. In the segmentation experiments on the RSDD dataset, compared with other known rail defect detection methods, the overall evaluation metric F-Score is 83.70 which is also better than other methods.

A One-Stage Approach For Surface Anomaly Detection With Background Suppression Strategies

AnomalySeg: Deep Learning-Based Fast Anomaly Segmentation Approach for Surface Defect Detection

An Attention-Based Network for Textured Surface Anomaly Detection

Unsupervised Anomaly Detection for Surface Defects with Dual-Siamese Network

A Two-Stage Anomaly Detection Framework: Towards Low Omission Rate in Industrial Vision Applications.

Industrial Product Surface Anomaly Detection with Realistic Synthetic Anomalies Based on Defect Map Prediction

Clear Memory-Augmented Auto-Encoder for Surface Defect Detection

SLSG: Industrial Image Anomaly Detection by Learning Better Feature Embeddings and One-Class Classification

Anomaly-GAN: A data augmentation method for train surface anomaly detection

An Adaptive Pyramid Graph and Variation Residual-Based Anomaly Detection Network for Rail Surface Defects

Focus Your Distribution: Coarse-to-Fine Non-Contrastive Learning for Anomaly Detection and Localization

Semantic Coarse-to-Fine Granularity Learning for Two-Stage Few-Shot Anomaly Detection

Autoencoder-Based Unsupervised Surface Defect Detection Using Two-Stage Training

AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios

Anomaly Detection Based on Data Super-Resolution in Industrial Cyber-Physical Systems with Multi-Rate Sampling

Progressive Boundary Guided Anomaly Synthesis for Industrial Anomaly Detection

Produce Once, Utilize Twice for Anomaly Detection

Normalizing Flow-Based Industrial Complex Background Anomaly Detection

Hyperspectral anomaly detection based on background suppression and adaptive threshold segmentation

AnomalySD: Few-Shot Multi-Class Anomaly Detection with Stable Diffusion Model

Industrial surface defect detection and localization using multi-scale information focusing and enhancement GANomaly