The excitation and vibration triggered by the long-term operation of railway vehicles inevitably result in defective states of catenary support devices. With the massive construction of high-speed electrified railways, automatic defect detection of diverse and plentiful fasteners on the catenary support device is of great significance for operation safety and cost reduction. Nowadays, the catenary support devices are periodically captured by the cameras mounted on the inspection vehicles during the night, but the inspection still mostly relies on human visual interpretation. To reduce the human involvement, this paper proposes a novel vision-based method that applies the deep convolutional neural networks (DCNNs) in the defect detection of the fasteners. Our system cascades three DCNN-based detection stages in a coarse-to-fine manner, including two detectors to sequentially localize the cantilever joints and their fasteners and a classifier to diagnose the fasteners' defects. Extensive experiments and comparisons of the defect detection of catenary support devices along the Wuhan-Guangzhou high-speed railway line indicate that the system can achieve a high detection rate with good adaptation and robustness in complex environments.

Split pins (SPs) play an important role in fixing joint components on catenary support devices (CSDs) of high-speed railway. The occurrence of loose and missing defects of SPs could make the structure of CSDs unstable. In this paper, we present a three-stage automatic defect inspection system for SPs mainly based on an improved deep convolutional neural network (CNN), which is called PVANET++. First, SPs are localized by PVANET++ and the Hough transform & Chan-Vese model, and then, three proposed criteria are applied to detect defects of SPs. In PVANET++, a new anchor mechanism is applied to produce suitable candidate boxes for objects, and multiple hidden layer features are combined to construct discriminative hyperfeatures. The performance of PVANET++ and several recent state-of-the-art deep CNNs is compared in a data set that is collected from a 60-km rail line. The results show that our model is superior to others in accuracy, and has a considerable speed.

The isoelectric line is an important component that connects the steady arm and the drop bracket of catenary in high-speed railway.An automatic fault detection system for the loose strands of isoelectric line is proposed.First, a CNN is adopted to extract the isoelectric line features. An improved feature extraction network, called as the isoelectric line network, is presented. Using the images captured from catenary inspection vehicles, the image areas that contain the isoelectric lines are obtained based on the Faster R-CNN. Second, the image segmentation is carried out based on the MRF model. And, the accurate isoelectric line pixels are obtained from the smallest image area extracted from the first stage. Lastly, the fault state is given by analyzing the quantity of the independent connection regions and the pixels standard deviation. Experimental results show that the proposed system has a high detection accuracy. Further, compared with the CNNs (VGG-16 and ZG) and typical detection method (HOG+SVM), the proposed network has better performance for the isoelectric line location.

The goal of this article is to develop a universal anomaly detection approach for catenary support components (CSCs) based on the generative adversarial networks (GANs). As the long-term operation of railway system, a wide range of failures, which threaten the safe operation of vehicles, perhaps happen to CSCs. Until now, it is hard to design a generic detection system to recognize all these kinds of failures because each defect needs a special detecting algorithm for different fault signatures. The lack of anomaly samples also makes it difficult for supervised learning methods to detect effectively. In this article, a novel approach, which combines deep convolution neural networks (DCNNs) with GANs, is proposed to estimate whether failures happen and give an alarm to stop the accident. First, an object location model is trained by DCNNs to obtain numerous samples of CSCs. Second, a generative model based on deep convolutional GAN (DCGAN) is constructed to find a good mapping from image space to high-dimensional feature spaces implicitly. Finally, an anomaly rating criterion is used to diagnose images. Two typical components of CSCs, the insulator that has big body and the isoelectric line that has tiny characters, are tested here. Experiments show that the proposed method can correctly judge anomalous images of CSCs and possess a good generic failure detection ability in this single framework.

The brace sleeve screw is an important component of catenary support devices in high-speed railways. According to the statistical data, there is a much risk of failure for brace sleeve screws during the operation of catenary system. In order to obtain better results of fault diagnosis for brace sleeve screws, it is of vital importance to develop a high-precision detection approach for brace sleeve screws. Aiming at the problems that the size of brace sleeve screw is too small and the previous method is difficult to detect accurately, a novel object detection network based on an improved Faster RCNN is proposed. Through the analysis of the factors that influence the detection accuracy for brace sleeve screws and the characteristics of original Faster RCNN, the concepts of discrimination maps and proposal maps are introduced to the improved Faster RCNN. In order to retain the characteristic information of brace sleeve screws, the high-resolution discrimination maps are adopted to identify the object type, and the low-resolution proposal maps are adopted to generate the candidate boxes of region with appropriate anchor scale configuration. Experimental results show that the proposed method has a higher detection accuracy for brace sleeve screws. In addition, compared with original Faster RCNN implementation based on VGG-16 and ResNet-101 backbone, the detection accuracy of the improved Faster RCNN is increased by 8%. It also provides a new idea for general object detection of small targets.

The catenary support components (CSCs) are the most important devices in high-speed railways to support contact lines for powering trains. To estimate the states of CSCs, it is very necessary to locate their positions in a monitoring system based on computer vision. Considering the application scenarios and characteristics of CSCs, an automatic and quick positioning system is designed in this paper to simultaneously position the multiscale CSCs with 12 categories. In the system, an effective framework called CSCs network (CSCNET) is presented, which cascades the coarse positioning network and the fine positioning network to reduce multiscale differences between different CSCs. In the coarse positioning network, a new unsupervised clustering algorithm based on the relative positioning information is proposed to classify the catenary images. Then, a convolutional neural network (CNN) classification network is trained to extract the structural features of catenary images and generate the proposal regions with labels. In the fine positioning network, a modified CNN positioning framework is applied to obtain the accurate positions of CSCs based on the coarse positioning results. Due to the special lightweight structure with a classification network, the relative position information is applied and makes the CSCNET sensitive to small-scale components. The experimental results from some high-speed railway lines in China show that the proposed system has obvious advantages in the CSCs positioning. The mean average precision and frames per second of CSCNET reach 0.837 and 2.17, respectively. Compared with some popular convolutional networks [faster region-based CNN (Faster R-CNN), etc.] and a typical positioning method, the proposed system significantly improves the AP without increasing the computational time.