Brace sleeve (BS) fasteners, i.e., nut and bolt, are small components but play essential roles in fixing BS and cantilever in railway catenary system. They are commonly inspected by onboard cameras using computer vision to ensure the safety of railway operation. However, most BS fasteners cannot be directly localized because they are too small in the inspection images. Instead, the BS is first localized for detecting the BS fastener. This leads to a new problem that the localized BS boxes may not contain the complete BS fasteners due to low localization accuracy, making it infeasible to further diagnose the fastener conditions. To tackle this problem, this article proposes a novel pipeline for BS fastener looseness diagnosis. First, the competitive deep learning model Faster RCNN ResNet101 is used to coarsely localize BSs. Second, an action-driven reinforcement learning agent is adopted to refine the coarse-localized boxes through a dynamic position searching process. Then, BS fasteners are extracted from the refined localized BS image by the deep segmentation model YOLACT++, which is fast and interpretable. Finally, a looseness diagnosis criterion based on segmented information are proposed. We evaluate the performance of submodels independently and the overall performance of the whole model on a real-life catenary image dataset collected from a high-speed line in China. The test results show that the proposed method is effective for BS looseness detection in railway catenary.

This article proposes an automatic high-precision detection method for structure parameters of catenary cantilever devices (SPCCDs) using 3-D point cloud data. The steps of the proposed detection method are: 1) segmenting and recognizing the components of the catenary cantilever devices, 2) extracting the detection plane and backbone component axis of catenary cantilever devices, and 3) detecting the SPCCD. The effective segmentation of components is critical for structure parameter detection. A point cloud segmentation and recognition method based on three-dimensional convolutional neural networks (3-D CNNs) is introduced to determine the different components of the catenary cantilever devices. Compared with traditional unsupervised clustering procedures for point cloud segmentation, the proposed method can improve the segmentation accuracy, does not require complex tuning procedures of parameters, and improves robustness and stability. Additionally, the segmentation method defines a recognition function, facilitating the analysis of the structural relationship between objects. Furthermore, we proposed an improved projection random sample consensus (RANSAC) method, which can effectively divide the detection plane of catenary cantilever devices to solve the multicantilever device occlusion problem. With RANSAC, it is also possible to precisely extract the backbone component axis and enhance parameter detection accuracy. The experimental results show that the structure angle and steady arm slope's error accuracy can achieve 0.1029° and 1.19%, respectively, which indicates the proposed approach can precisely detect the SPCCD.

The development of algorithms for detecting failures in railway catenary support components has, among others, one major challenge: data about healthy components are much more abundant than data about defective components. In this paper, virtual reality technology is employed to control the learning environment of convolutional neural networks (CNNs) for the automatic multicamera-based monitoring of catenary support components. First, 3D image data based on drawings and real-life video images are developed. Then, a virtual reality environment for monitoring the catenary support system is created, emulating real-life conditions such as measurement noise and a multicamera train simulation to resemble state-of-the-art monitoring systems. Then, CNNs are used to extract and fuse the features of multicamera images. Experiments are conducted for monitoring the cantilever support connection, both down (CSC-D) and up (CSC-U), and registration arm support connection, both down (RASC-D) and up (RASC-U). Experimental results show that the CNNs trained in the virtual reality environment can capture the most relevant spatial information of the catenary support components. Multicamera image detection based on CNNs detects screw loss for all four components. For CSC-D and RASC-U, normal and pin-loss images are also fully detected. A challenge remains in increasing the pin-loss detection for both CSC-U and RASC-D.