Visual Fault Detection of Multi-scale Key Components in Freight Trains

TL;DR

This paper introduces a lightweight, anchor-free deep learning framework for fault detection in freight train components, effectively handling multi-scale parts with high accuracy and low resource requirements.

Contribution

It proposes a novel lightweight, anchor-free model with a feature pyramid network to improve multi-scale fault detection accuracy in freight trains.

Findings

Achieves 98.44% detection accuracy

Model size is only 22.5 MB

Outperforms existing state-of-the-art detectors

Abstract

Fault detection for key components in the braking system of freight trains is critical for ensuring railway transportation safety. Despite the frequently employed methods based on deep learning, these fault detectors are highly reliant on hardware resources and are complex to implement. In addition, no train fault detectors consider the drop in accuracy induced by scale variation of fault parts. This paper proposes a lightweight anchor-free framework to solve the above problems. Specifically, to reduce the amount of computation and model size, we introduce a lightweight backbone and adopt an anchor-free method for localization and regression. To improve detection accuracy for multi-scale parts, we design a feature pyramid network to generate rectangular layers of different sizes to map parts with similar aspect ratios. Experiments on four fault datasets show that our framework achieves 98.44% accuracy while the model size is only 22.5 MB, outperforming state-of-the-art detectors.