Search

This article aims to improve the performance of deep learning-based defective fastener inspection method. Due to the defective fasteners are insufficient and far less than the defect-free ones in real railway, it is difficult to train a robust fastener inspection model on such imbalanced dataset. In view of this problem, a novel image generation method called four-discriminator cycle-consistent adversarial network (FD-Cycle-GAN) is proposed to generate the defect fastener images using a large number of defect-free ones. Extensive experiments are conducted on the real fastener images and generated images. Experimental results demonstrate that the defect fastener images generated by our proposed method have better quality and richer diversity than those generated by other state-of-the-art methods. In addition, compared with the CNN-only baseline, the performance of the fastener inspection model trained on the expanded dataset containing the defect fastener images generated by FD-Cycle-GAN is improved significantly. The detection accuracy and relative IMP reach 93.25% and 21.59% respectively. [ABSTRACT FROM AUTHOR]

Visual inspection techniques for rail surface defects have become prevalent approaches to obtain information on rail surface damage. However, uneven illumination leads to illegibility of local information, and the change of the wheel-rail area results in the changeful background of the rail surface, both of which pose challenges to the visual inspection. This paper proposes a novel algorithm that detects rail surface defects via partitioned edge features (PEF). PEF eliminates the effect of uneven illumination by effectively extracting edge features and building homogeneous background on the rail surface. In the process of edge feature extraction, the thresholding based on adaptive partition of rail surface (APRS) plays an indispensable role. In APRS, the rail surface is adaptively partitioned into three types of regions according to the wheel-rail contact degree. After that, the dynamic threshold is set adaptively for each region type on the basis of the prior information of defect proportion. Subsequently, based on neighborhood information and fuzzy decision, the spatial information of adjacent pixels and the direction information of fracture edges are utilized to realize the effective recovery of incomplete defect contours. In addition, defect contours are precisely filled via a flexible combination of morphological hole filling operation and defect region extraction based on improved background difference. The accuracy of this PEF algorithm was confirmed by experiments and comparisons with related algorithms. The experiment results show that PEF detects defects with 92.03% recall and 88.49% precision, which achieves higher accuracy than the established detection algorithms for rail surface defects. [ABSTRACT FROM AUTHOR]

Rail surface defect inspection based on machine vision faces challenges against the complex background with interference and severe data imbalance. To meet these challenges, in this article, we regard defect detection as a key-point estimation problem and present the proposed attention neural network for rail surface defect detection via consistency of Intersection-over-Union(IoU)-guided center-point estimation (CCEANN). The CCEANN contains two crucial components. The two components are the stacked attention Hourglass backbone via cross-stage fusion of multiscale features (CSFA-Hourglass) and the CASIoU-guided center-point estimation head module (CASIoU-CEHM). Furthermore, the CASIoU-guided center-point estimation head module integrating the delicate coordinate compensation mechanism regresses detection boxes flexibly to adapt to defects’ large-scale variation, in which the proposed CASIoU loss, a loss regressing the consistency of intersection-over-union (IoU), central-point distance, area ratio, and scale ratio between the targeted defect and the predicted defect, achieves higher regression accuracy than state-of-the-art IoU-based losses. The experiments demonstrate that the CCEANN outperforms competitive deep learning-based methods in four surface defect datasets. [ABSTRACT FROM AUTHOR]

Rail corrugation (RC) is one of the most important indicator to evaluate the quality of rail, which is used to describe the irregularity of rail surface, also known as rail’s vertical flatness. However, the definition of RC is still an empirical description for low-speed measurement devices. In this paper, the process of RC measurement is divided into two steps, sampling and extraction, which helps the users to understand more clearly. Then a new mathematic model is proposed to make the process of RC’s extraction be an executable operation for machine calculation. The model adopts a new concept of flexible virtual ruler to perform sliding filtering on an overall rail and extract the instantaneous RC with a successive approximation algorithm according to user’s requirements and national standards. The proposed FVR model can not only be fully compatible with the traditional extraction method of RC, but also provide a new idea for evaluating RC’s quality. Comparing with the current popular methods, the proposed model gives a meaningful strategy for rail maintenance with a complete mathematic description having more degrees of freedom. Experiment results demonstrate its validity and reliability for both indoor simulation and actual outdoor experiments. [ABSTRACT FROM AUTHOR]

Rail corrugation is the geometric structure of the rail top surface, and it directly related to the safety and comfort of rail transportation. The monitoring of rail corrugation is the key work for rail maintenance department. Currently, the 2D laser displacement sensor is being used to measure the rail corrugation. However, the 2D sensor is easily affected by pitching vibration under complex measurement environment, which leads to large errors in the final measurement results. For solving the problem of pitching vibration in dynamic measuring rail corrugation, a data splicing method based on deviation recognition and calibration is proposed in this study. Firstly, the blind recognition of sensor deviation is applied to identify the measurement state of sensor in each data sampling. Then, a mapping method is proposed to position the actual repeated measurement interval according to geometric relationship between the deviation angle and the offset in the longitudinal direction. Finally, the modified DTW model is proposed for data registration and the data splicing can be achieved according to the registration results. Extensive experiments are carried on both indoor experimental platform and outdoor platform. The experimental results indicate that our proposed method can accurately recognize the deviation of sensor and can calibrate the deviated data set, and its performance on the rail corrugation measurement under the pitching vibration case outperforms previous data splicing work. Thus, it is with the potential engineering application. [ABSTRACT FROM AUTHOR]

For rail wear measurement, the process of profile registration is critical. Nowadays, the widely used method is the rail waist double circle segment method (DCS). However, this method may be invalid in actual applications because of the influence of the following two factors: one is the outliers mixed into profiles and the other is the profile diversity. The former induces the misalignment between the measured profile and the standard one, and the latter may destroy the integrity of rail waist, which makes that the measured profile cannot be matched with the standard one through DCS. To solve the two problems, a hybrid profile registration method is proposed in this paper. First, by locating new matching primitives to realize coarse registration, we check the profile integrity fast. Then, for complete profile, we rematch its original measured profile with the standard one finely based on DCS to detect and remove the outliers correctly. The efficiency and superiority of the proposed method were verified by numerous experiments. The results show that the average score of F1-Measure for outlier detection reaches 0.95, which outperforms some classical models obviously. Meanwhile, the result of the profile integrity check is also basically coincided with the real face of test zones. Furthermore, the system can run at a speed of 21.95 km/h under our experimental setup, which is far higher than that of the rail maintenance train (up to 5 km/h). [ABSTRACT FROM AUTHOR]

Fasteners are an important part of railway security. The application of computer vision and image processing on fastener detection has become a research hot spot. However, most of the researches have focused on the missing and damage detection of fasteners up to now. In this paper, a method based on image processing is proposed to locate the center of the hexagon nut in the railway fastener. The precise position of the hexagon nut can be used for the automatic replacement of the nut, which is of great significance for the automatic railway construction and maintenance. The locating method proposed in this paper can be divided into two steps. First, the improved Canny detector which combines surrounding suppression and iterative excitation is used to extract the contours of the fastener. Then, a multifeature hierarchical locating (MHL) algorithm is proposed, which integrates the circle contour of the gasket and the hexagon features of the nut. The experimental results show that the improved Canny detector suppresses noises effectively and highlights the contours of fastener. The MHL algorithm in the next step improves the locating accuracy remarkably while guaranteeing the robustness meantime. [ABSTRACT FROM AUTHOR]

Rail corrugation dynamic measurement techniques are critical to guarantee transport security and guide rail maintenance. The chord-based method is one of the most common methods nowadays. However, for corrugation broadband measurement varied from 30 mm to 60 m, this method is faced with two severe challenges in actual applications: one is that the effective measuring range of traditional single chord models cannot cover the required broadband, and the other is that the common frequency sampling method (FSM) is unavailable to design the FIR inverse filters for asymmetrical chord and multipoints chord models. To solve the two problems, a practical combination-chord model (CCM) is proposed in this paper. For the first problem, the model is constructed on the existing three-point asymmetrical chord platform. Through the flexible combination of three sensors, CCM integrates the advantages of long-wave measurement from the two-point chord model and shortwave measurement from the three-point asymmetrical chord model, and realizes the corrugation broadband measurement in one platform successfully. For the second problem, the least square (LS) is employed to design the inverse filter. Compared with FSM, the design process is more unlimited and easier for the user. More important, even with the same output filter order, the filter performance designed by LS is better than that by FSM. The efficiency of CCM and the superiority of inverse filter designed by LS were verified by vast experiments. The results could achieve the design target. [ABSTRACT FROM PUBLISHER]

Dynamically measuring rail profile using the structured-light vision suffers from random vibrations on the line laser projector, which would cause distorted rail profiles. This paper presents a simple and effective distortion rectifying method to address this issue. The distorted rail profile is rectified by easily projecting it onto an auxiliary plane which is parallel to the cross section of rail. In order to establish the auxiliary plane, three profiles formed by radiating multiline structured light on rail are used to fit the rail longitudinal axis. More importantly, only one of the light planes is required to be calibrated beforehand. The others are calibrated online with the proposed self-calibration method, which is based on collinearity and parallelity constraints on the scene points of different rail profiles and requires only one image of the scene. Apart from evaluating the implementation with comprehensive experiments, we compare our method against other published works. The results exhibit its effectiveness and superiority in terms of the dynamic measurement of the rail profile. [ABSTRACT FROM AUTHOR]

Rail section profile accurate measurement techniques are critical to guarantee the transport security and reduce the maintenance cost. However, the profile may make an affine transformation under vehicle vibration. It is a challenge to identify and calibrate the affine distorted profiles in the 2-D laser displacement system (LDS). By profound observation and discussion on triple primitives including rail jaw, railhead inner line, and rail waist curve, we found the difference between normal profiles and distorted ones. To measure the difference quantitatively, we proposed a novel curve similarity measurement method, i.e., point set mapping (PSM), in this paper. By constructing PSM from the standard profile to the measured one on the triple matching primitives, the fast recognition and accurate calibration of rail profile under affine distortion were implemented. The accuracy and efficiency of the system were verified by experiments and comparisons with related methods. The results show that the PSM is faster and more robust than the universal Hausdorff distance on the similarity measure, and the performance of the modified particle swarm optimization is superior for profile distortion calibration. Furthermore, the LDS can run at a speed of 12.86 km/h under our experimental setup, which is far higher than that of the rail maintenance train (up to 5 km/h). [ABSTRACT FROM PUBLISHER]