Peng Dai

The share of rail transport in world transport continues to rise. As the number of trains increases, so does the load on the railway. The rails are in direct contact with the loaded wheels. Therefore, it is more easily damaged. In recent years, domestic and foreign scholars have conducted in-depth research on railway damage detection. As the weakest part of the track system, switches are more prone to damage. Assessing and predicting rail surface damage can improve the safety of rail operations and allow for proper planning and maintenance to reduce capital expenditure and increase operational efficiency. Under the premise that functional safety is paramount, predicting the service life of rails, especially turnouts, can significantly reduce costs and ensure the safety of railway transportation. This paper understands the evolution of contact fatigue on crossing noses through long-term observation and sampling of crossing noses in turnouts. The authors get images from new to damaged. After image preprocessing, MPI (Magnetic Particle Imaging) is divided into blocks containing local crack information. The obtained local texture information is used for regression prediction using machine-supervised learning and LSTM network (Long Short-Term Memory) methods. Finally, a technique capable of thoroughly evaluating the wear process of crossing noses is proposed.

Image-based rail defect detection could be conceptually defined as an object detection task in computer vision. However, unlike academic object detection tasks, this practical industrial application suffers from two unique challenges, including object ambiguity and insufficient annotations. To overcome these challenges, we introduce the pixel-wise attention mechanism to fully exploit features of annotated defects, and develop a feature augmentation framework to tackle the defect detection problem. The pixel-wise attention is conducted through a learnable pixel-level similarity between input and support features to obtain augmented features. These augmented features contain co-existing information from input images and multi-class support defects. The final output features are augmented and refined by support features, thus endowing the model to distinguish between ambiguous defect patterns based on insufficient annotated samples. Experiments on the rail defect dataset demonstrate that feature augmentation can help balance the sensitivity and robustness of the model. On our collected dataset with eight defected classes, our algorithm achieves 11.32% higher mAP@.5 compared with original YOLOv5 and 4.27% higher mAP@.5 compared with Faster R-CNN.

The detection of rail surface defects is vital for high-speed rail maintenance and management. The CNN-based computer vision approach has been proved to be a strong detection tool widely used in various industrial scenarios. However, the CNN-based detection models are diverse from each other in performance, and most of them require sufficient training samples to achieve high detection performance. Selecting an appropriate model and tuning it with insufficient annotated rail defect images is time-consuming and tedious. To overcome this challenge, motivated by ensemble learning that uses multiple learning algorithms to obtain better predictive performance, we develop an ensemble framework for industrialized rail defect detection. We apply multiple backbone networks individually to obtain features, and mix them in a binary format to obtain better and more diverse sub-networks. Image augmentation and feature augmentation operations are randomly applied to further make the model more diverse. A shared feature pyramid network is adopted to reduce model parameters as well as computation cost. Experimental results substantiate that the approach outperforms single detecting architecture in our specified rail defect task. On the collected dataset with 8 defect classes, our algorithm achieves 7.4% higher mAP.5 compared with YOLOv5 and 2.8% higher mAP.5 compared with Faster R-CNN.