LAUSR

The pantograph–catenary system is crucial for transferring electrical power from catenary lines to electrified train and the occurrence of arcing could damage railway operations; thus, it is important to detect arcing. Detecting arcing in complex scenes and detecting different sizes and shapes of arcing is still a challenge. To overcome these issues, a robust image-based semantic segmentation model named arc multiscale fusion (ArcMSF) is proposed for arcing detection of pantograph–catenary, which designs a novel hybrid multiscale feature fusion model that aggregates Transformer with convolutional neural network (CNN) to realize arcing pixel segmentation. A down-top decoder for combining low-level features with high-level features is designed to achieve multiscale-level arcing feature detection in complex scenes. Inspired by the arcing image properties that arcing is always the brightest, the global max features and global threshold features are designed to augment the arcing features. Experiments on IVAIS-PCA2021 dataset and comparative experiments are conducted to demonstrate the effectiveness of the ArcMSF, which can achieve an 89.13% segmentation accuracy and a fast inference speed of 23.84 ms. Moreover, the detection results have a clear depiction of edge details.