Supervised Anomaly Detection Method Combining Generative Adversarial Networks and Three-Dimensional Data in Vehicle Inspections

TL;DR

This paper introduces a supervised anomaly detection approach using GAN-based style conversion on 3D graphics to generate training data for vehicle inspection, improving detection accuracy without extensive anomaly data.

Contribution

The study presents a novel supervised method combining GANs and 3D graphics to generate anomaly images for training, addressing data scarcity in vehicle inspection anomaly detection.

Findings

Successfully generated realistic anomaly images with style conversion.

Achieved effective anomaly detection using generated supervised data.

Reduced need for complex data collection and annotation.

Abstract

The external visual inspections of rolling stock's underfloor equipment are currently being performed via human visual inspection. In this study, we attempt to partly automate visual inspection by investigating anomaly inspection algorithms that use image processing technology. As the railroad maintenance studies tend to have little anomaly data, unsupervised learning methods are usually preferred for anomaly detection; however, training cost and accuracy is still a challenge. Additionally, a researcher created anomalous images from normal images by adding noise, etc., but the anomalous targeted in this study is the rotation of piping cocks that was difficult to create using noise. Therefore, in this study, we propose a new method that uses style conversion via generative adversarial networks on three-dimensional computer graphics and imitates anomaly images to apply anomaly detection based on supervised learning. The geometry-consistent style conversion model was used to convert the image, and because of this the color and texture of the image were successfully made to imitate the real image while maintaining the anomalous shape. Using the generated anomaly images as supervised data, the anomaly detection model can be easily trained without complex adjustments and successfully detects anomalies.