RSDNet: A New Multiscale Rail Surface Defect Detection Model.

The speed and load capacity of trains are improving at an accelerating rate, which raises the standards for railway services' security criteria. The track's surface may progressively reveal varying degrees of defects because of the impacts of moisture, temperature, load, and other factors; if the flaws never addressed in a prompt period, the degree of defects would expand, significantly raising the probability of train service. The development of automatic RSDD (Rail Surface Defect Detection) has substantial practical and scientific implications. In this study, a multi-crack detection technique depends on Variational Autoencoder Diffusion Model (VAEDM), The introduction of VAEDM allows for the non-destructive detection of fastener and RSD. In order to produce hidden elements from images that have various cracks, VAEDM combines the multi-layer and a rail surface image encoder was introduced. Zero Shot- Divergence Asynchronous Reinforcement Learning (ZS-DARL), Policy-Gradient is used to establish the relationship between defects and non-defects. DARL transformer encoder is introduced for taking long-range dependences for FE (Feature Extraction) from detected objects images with various patterns and dimensions. The simulation outcomes on the open Railway Track Fault Detection (RTFD) and Rail Surface Defect Datasets (RSDDs) with rail surface defects are collected from rail tracks surface defect detection. Results are measured using the metrics like recall, precision, F-measure, and accuracy.

Due to the rapid advances in railway industry, the rail surface defect detection task which inspects whether the rail is defective has become an increasingly critical issue. Detecting rails by an automatic and swift approach instead of present manual inspections enables the work more efficient and safe currently. In this paper, we propose a novel two-stage pipeline method for rail defect detection by localizing and classifying rail images. Specifically, in the first stage, we get the cropped images which focus on the rail part instead of the whole-original image by integrating traditional image processing methods. In the second stage, we put the cropped images into a fine-tuned convolution neural network (CNN) and extract part-level features for rail images classification. Especially, in the rail image defect detection scenario, we should take recall into account to some extent, so we propose a novel loss function to leverage both of them in the second stage. The results show that the proposed method has strong robustness and achieves practical performance in defect detection precision.

With the rapid development of rail transit, the workload of track maintenance has increased, making the intelligent identification of rail surface defects crucial for improving detection efficiency. To address issues such as low defect detection accuracy, the loss of feature information due to single-path architecture backbones, and insufficient information interaction in existing rail defect detection methods, we propose a rail surface defect detection method based on dual-path feature fusion (DPF). This method initially employs a dual-path structure to separately extract low-level and high-level features. It then utilizes a combination of attention mechanisms and feature fusion techniques to integrate these features. By doing so, it preserves richer information and enhances detection accuracy and robustness. The experimental results demonstrate that the comprehensive performance of the proposed model is superior to mainstream algorithms.

Rail surface defect detection plays a critical role in the maintenance of the rail transportation system. Video analysis technology is a promising method to detect defects due to its low cost and effectiveness. Several attempts with hand‐craft features have been made to obtain the detection results by using traditional machine vision algorithms. However, these methods suffer from imprecise results due to challenging conditions, such as deteriorated and changeable lighting environment and various types of complex rail surface defects. Recently, classification methods with complex deep convolutional networks have become popular. Despite their high accuracy, these methods cannot meet the requirements of defects localisation and real‐time processing in practice. To solve these problems, this study proposes a novel object detection algorithm to detect rail defects. The net architecture of the proposed algorithm includes a backbone network using MobileNet and several novel detection layers with multi‐scale feature maps inspired by you only look once (YOLO) and feature pyramid networks. Two different architectures of MobileNet are used to estimate the performance of defects detection. The experimental results demonstrate the great potential of the proposed algorithm with fast inference speed and high accuracy in the industry.

This paper proposes a multi-source data fusion algorithm for rail surface defect detection in both camera-based rail inspection images and ultrasound B-scan images. First, we design a rail surface segmentation algorithm based on image bilateral filtering, Sobel edge detection, and rail surface edge detection to extract the rail surface area. Second, we build a camera and ultrasound data fusion (CUFuse) model for rail surface defect detection, including two main networks: multi-source data feature extraction and multi-scale feature fusion networks. The multi-source data feature extraction network consists of two BoTNet 50 networks as feature extraction networks to extract five stages of features in camera-based images and ultrasound B-scan images. The multi-scale feature fusion network consists of five feature fusion modules to fuse the feature information output by the multi-source data feature extraction network. Finally, we use the CUFuse model to detect the rail surface defect dataset, and output five rail surface state types, including Light, Moderate, Severe, Normal, and Joint. The results show that the accuracy of the CUFuse model is 96.97%, which can accomplish the task of rail surface defect detection on railway sites.

As track components deteriorate, their interaction dynamic response will alter accelerations on the train. Measuring acceleration on train components is a method to conduct condition monitoring of track defects. The measured signal can be used for the detection of rail surface defects which implement impacts on measured accelerations. In this paper, the axle-box acceleration (ABA) is measured in a subway as a case study. Fourier transform (FFT), empirical mode decomposition (EMD) and ensemble EMD (EEMD) methods are used to study accelerations relative to the track. Wheel frequencies are calculated using finite element method (FEM) to determine frequency couplings. Velocity-dependent/independent components and source of excitation of the measured signal are distinguished. Results indicated that the FFT approach can be applied for both velocity-dependent and velocity-independent vibration components for frequencies up to 680 Hz. Also, the EEMD method can be used to distinguish the impact component of the measured signal.

The detection of rail surface defects is an important tool to ensure the safe operation of rail transit. Due to the complex diversity of track surface defect features and the small size of the defect area, it is difficult to obtain satisfying detection results by traditional machine vision methods. The existing deep learning-based methods have the problems of large model sizes, excessive parameters, low accuracy and slow speed. Therefore, this paper proposes a new method based on an improved YOLOv4 (You Only Look Once, YOLO) for railway surface defect detection. In this method, MobileNetv3 is used as the backbone network of YOLOv4 to extract image features, and at the same time, deep separable convolution is applied on the PANet layer in YOLOv4, which realizes the lightweight network and real-time detection of the railway surface. The test results show that, compared with YOLOv4, the study can reduce the amount of the parameters by 78.04%, speed up the detection by 10.36 frames per second and decrease the model volume by 78%. Compared with other methods, the proposed method can achieve a higher detection accuracy, making it suitable for the fast and accurate detection of railway surface defects.

To address the challenges of accurately extracting features from elongated scratches, irregular defects, and small-scale surface flaws on high-precision linear guide rails, this paper proposes a novel instance segmentation algorithm tailored for guide rail surface defect detection. The algorithm integrates the YOLOv8 instance segmentation framework with deformable convolutional networks and multi-scale feature fusion to enhance defect feature extraction and segmentation performance. A dedicated guide rail surface Defect (GSD) segmentation dataset is constructed to support model training and evaluation. In the backbone, the DCNv3 module is incorporated to strengthen the extraction of elongated and irregular defect features while simultaneously reducing model parameters. In the feature fusion network, a multi-scale feature fusion module and a triple-feature encoding module are introduced to jointly capture global contextual information and preserve fine-grained local defect details. Furthermore, a Channel and Position Attention Module (CPAM) is employed to integrate global and local features, improving the model’s sensitivity to channel and positional cues of small-target defects and thereby enhancing segmentation accuracy. Experimental results show that, compared with the original YOLOv8n-Seg, the proposed method achieves improvements of 3.9% and 3.8% in Box and Mask mAP50, while maintaining a real-time inference speed of 148 FPS. Additional evaluations on the public MSD dataset further demonstrate the model’s strong versatility and robustness.

As an important component of the railway system, the surface damage that occurs on the rails due to daily operations can pose significant safety hazards. This paper proposes a simple yet effective rail surface defect detection model, FS-RSDD, for rail surface condition monitoring, which also aims to address the issue of insufficient defect samples faced by previous detection models. The model utilizes a pre-trained model to extract deep features of both normal rail samples and defect samples. Subsequently, an unsupervised learning method is employed to learn feature distributions and obtain a feature prototype memory bank. Using prototype learning techniques, FS-RSDD estimates the probability of a test sample belonging to a defect at each pixel based on the prototype memory bank. This approach overcomes the limitations of deep learning algorithms based on supervised learning techniques, which often suffer from insufficient training samples and low credibility in validation. FS-RSDD achieves high accuracy in defect detection and localization with only a small number of defect samples used for training. Surpassing benchmarked few-shot industrial defect detection algorithms, FS-RSDD achieves an ROC of 95.2% and 99.1% on RSDDS Type-I and Type-II rail defect data, respectively, and is on par with state-of-the-art unsupervised anomaly detection algorithms.

Heavy-haul railways play a vital role in freight transportation, and the health of the rails directly impacts the safety and efficiency of railway operations. The heavy axle loads and long train compositions of heavy-haul trains make the rail surface susceptible to damage such as rail corrugation, spalling and abrasion, threatening operational safety. To address the issue, this paper proposes a multi-source data fusion method for identifying rail surface defects on heavy-haul railways. First, complete ensemble empirical mode decomposition with adaptive noise is used to decompose vibration signals and extract multi-dimensional vibration features. Next, dynamic time warping is applied to align rail profile data and extract key geometric features. Then, the vibration features and profile features are fused using Relief-F to select the most discriminative features. Finally, a support vector machine is utilized for defect identification. Experiment results show that the proposed method achieves high accuracy in identifying rail surface defects, with an accuracy of 96.4%.

The application of computer vision technology in defect detection of industrial products is a popular research direction in recent years. This article presents the pyramid feature convolutional neural network (CNN) for defect detection of rail surfaces. First, multi-scale feature maps are extracted based on the characteristics of defects and backgrounds by the pyramid feature extraction module (PFEM). Then the feature maps are input to a lightweight network consisting of a small number of parameters. The network is trained with only 40% data of the dataset using binary cross-entropy loss function and the intersection of union (IOU) loss function. In the experiment, the performance of the proposed method is evaluated using the rail surface defect dataset (RSDD) dataset by comparing it with other methods. The experimental results show that the segmentation performance and real-time performance of the proposed method are better than those of other methods.

Surface defects are usually the early phenomenon of rail failure, which threatens the safety of railroad transportation critically, and the timely detection of surface defects helps to eliminate the potential risk of rail and reduce the chance of railroad safety accidents. The existing methods of detecting surface defects on rails suffer from a large performance degradation in the application of rails containing pollutions such as rust and oil. Therefore, this article proposes a multilevel, end-to-end fast rail surface defect detection method. First, rail extraction was performed based on the stability of the standard deviation of the edge pixels. Then, differential box-counting (DBC) and GrabCut algorithm are then combined for defect segmentation to boost the speed and accuracy of extracting complex defects. Finally, YOLO v2 is used to precisely locate and detect defects. The experimental results show that the proposed method performs well, with an average accuracy of 97.11%, an average recall of 96.10%, and an average frame rate of 0.0064 s. In addition, the proposed method offers a high robustness in the tests of different use cases.

Rail track surfaces could suffer from defects such as abrasion and deformation, thus to ensure rail system safety, the conditions of rail tracks must be monitored. With the advancement of deep learning and computer vision technologies, automatic detection and classification techniques are being tested to replace or complement manual patrolling for productivity improvement. However, classic neural network approaches require a large amount of data which could be time-consuming and limit the application of deep learning techniques. This paper proposes applying one-shot learning using a Siamese convolutional neural network to the identification of rail surface defects. The results show the reduced requirement of training speed and possess potentials for real-time applications.KeywordsRail surface defectDefect classificationOne-shot learningSiamese network

Defects on the surface of railroad tracks have been the cause of growing concern over the past three decades. The automated detection and classification of rail surface defects would be of great assistance to rail maintenance planners, who develop grinding strategies to prevent the development of potentially dangerous deterioration. Videotaped images of the surface of rail have been obtained, but they are subject to distortions due to the acquisition process as well as physical phenomena on the track itself. In this analysis, an algorithm is presented for the simultaneous restoration and segmentation of objects in a two-dimensional image. The algorithm relies on distributions that model the relationships between sites and neighbors in order to restore a distorted image to an estimate of its ideal form, and also obtain detailed information about the objects located in the image. The foundation of the algorithm is the Iterated Conditional Modes procedure for image restoration. The resulting extension is capable of providing detailed measurements of the geometric features of objects detected in an image. The extended algorithm is applied to an image distorted by simulated noise, and also to an image taken from a videotape of a rail surface. The results of the analysis demonstrate the potential for accurate detection, measurement, and classification of rail surface defects. Key Words: Image processingIterated conditional modesRestoration

Surface defect identification is a crucial task in many manufacturing systems, including automotive, aircraft, steel rolling, and precast concrete. Although image-based surface defect identification methods have been proposed, these methods usually have two limitations: images may lose partial information, such as depths of surface defects, and their precision is vulnerable to many factors, such as the inspection angle, light, color, noise, etc. Given that a three-dimensional (3D) point cloud can precisely represent the multidimensional structure of surface defects, we aim to detect and classify surface defects using a 3D point cloud. This has two major challenges: (i) the defects are often sparsely distributed over the surface, which makes their features prone to be hidden by the normal surface and (ii) different permutations and transformations of 3D point cloud may represent the same surface, so the proposed model needs to be permutation and transformation invariant. In this paper, a two-step surface defect identification approach is developed to investigate the defects’ patterns in 3D point cloud data. The proposed approach consists of an unsupervised method for defect detection and a multi-view deep learning model for defect classification, which can keep track of the features from both defective and non-defective regions. We prove that the proposed approach is invariant to different permutations and transformations. Two case studies are conducted for defect identification on the surfaces of synthetic aircraft fuselage and the real precast concrete specimen, respectively. The results show that our approach receives the best defect detection and classification accuracy compared with other benchmark methods.