Search Results (72)

Detecting metal surface crack defects is of great significance for the safe operation of industrial equipment. However, most existing mainstream deep-object detection models suffer from complex structures, large parameter sizes, and high training costs, which hinder their deployment and application in frontline construction sites. Therefore, this paper optimizes the existing YOLO series head structure and proposes a lightweight detection head structure, ZoomHead, with lower computational complexity and stronger detection performance. First, the GroupNorm2d module replaces the BatchNorm2d module to stabilize the model’s feature distribution and accelerate the training speed. Second, Detail Enhanced Convolution (DEConv) replaces traditional convolution kernels, and shared convolution is adopted to reduce redundant structures, which enhances the ability to capture details and improves the detection performance for small objects. Next, the Zoom scale factor is introduced to achieve proportional scaling of the convolution kernels in the regression branch, minimizing redundant computational complexity. Finally, using the YOLOv10 and YOLO11 series models as baseline models, ZoomHead was used to replace the head structure of the baseline models entirely, and a series of performance comparison experiments were conducted on the rail surface crack dataset and NEU surface defect database. The results showed that the integration of ZoomHead effectively improved the model’s detection accuracy, reduced the number of parameters and computations, and increased the FPS, achieving a good balance between detection accuracy and speed. In the comparative experiment of the SOTA model, the addition of ZoomHead resulted in the model having the smallest number of parameters and the highest FPS, while maintaining the same mAP value as the SOTA model, indicating that the ZoomHead structure proposed in this paper has better comprehensive detection performance. Full article

Rail surface defects pose significant risks to the operational efficiency and safety of industrial equipment. Traditional visual defect detection methods typically rely on high-quality RGB images; however, they struggle in low-light conditions due to small, low-contrast defects that blend into complex backgrounds. Therefore, this paper proposes a novel defect segmentation method leveraging a dual-stream feature fusion network that combines polarization images with DeepLabV3+. The approach utilizes the pruned MobileNetV3 as the backbone network, incorporating a coordinate attention mechanism for feature extraction. This reduces the number of model parameters and enhances computational efficiency. The dual-stream module implements cascade and addition strategies to effectively merge shallow and deep features from both the original and polarization images. This enhances the detection of low-contrast defects in complex backgrounds. Furthermore, the CBAM is integrated into the decoding area to refine feature fusion and mitigate the issue of missing small-target defects. Experimental results demonstrate that the enhanced DeepLabV3+ model outperforms existing models such as U-Net, PSPNet, and the original DeepLabV3+ in terms of MIoU and MPA metrics, achieving 73.00% and 80.59%, respectively. The comprehensive detection accuracy reaches 97.82%, meeting the demanding requirements for effective rail surface defect detection. Full article

Heavy rolling contact fatigue (RCF) may be caused by wheel surface defects under the influence of rail corrosion, which threatens the operational safety of rail vehicles. To investigate the role of surface defects on wheel RCF damage under the influence of rail corrosion, a salt spray tester was used to corrode the rails, an impact testing machine was employed to create surface defects, and RCF tests were completed. The role of surface defects on wheel RCF damage was studied by monitoring the wheel defect surface and cross-section. The results indicate that the tendencies of the RCF crack extension of surface defects of different sizes are similar, and they all extend in a C-shape along the tangential force direction. However, the larger the defect size, the later the crack is initiated. The leading edge material is continuously squeezed into the defect by the tangential force, and a larger plastic deformation layer is formed, which causes the RCF at the leading edge to crack more severely. Meanwhile, under the effect of combined normal force and shear stress, the leading edge crack intersects with the middle edge crack, and the leading edge material is spalled off first. Wheel RCF damage and wear are aggravated by rail corrosion, the longer the corrosion time, the more serious the RCF damage and wear, and the earlier the material spalling time, the lower the fatigue life. Full article

With the vigorous development of speedy railway technology, steel rails, as an important structural infrastructure in speedy railways, play a crucial role in ensuring the safety of the entire speedy railway operation. A brand-new type of speedy rail inspection robot and its fault detection method are proposed to solve a number of problems, such as the difficulty and low accuracy of real-time online detection of rail defects and damage in speedy railways. The brand-new rail inspection robot is driven by two drive wheels and adopts a standard rail gauge of 1435 mm, which ensures its speedy and smooth operation on the track as well as accurate motion posture information. Firstly, 12 common types of surface damage of the rail head were analyzed and classified into five categories based on their damage characteristics. The motion state of the rail inspection robot under the five types of surface damage of the rail head was analyzed and subjected to kinematic analysis. This study analyzed the relationship between the distinctive types of damage and the motion posture of the robot during the inspection of the five types of damage. Finally, experimental tests were conducted, and it was found that the robot’s motion posture would undergo sudden changes when inspecting distinctive types of injuries; the highest error rate was 3%. The effectiveness of this method was verified through experiments, and the proposed new track detection robot can greatly improve the track detection efficiency of high-speed railways and has specific academic research value and practical application value. Full article

To address the challenges in existing multi-sensor data fusion methods for rail surface defect diagnosis, particularly their limitations in fully exploiting potential synergistic information among multimodal data and effectively bridging the semantic gap between heterogeneous multi-source data, this paper proposes a diagnostic approach based on a Progressive Joint Representation Graph Attention Fusion Network (PJR-GAFN). The methodology comprises five principal phases: Firstly, shared and specific autoencoders are used to extract joint representations of multimodal features through shared and modality-specific representations. Secondly, a squeeze-and-excitation module is implemented to amplify defect-related features while suppressing non-essential characteristics. Thirdly, a progressive fusion module is introduced to comprehensively utilize cross-modal synergistic information during feature extraction. Fourthly, a source domain classifier and domain discriminator are employed to capture modality-invariant features across different modalities. Finally, the spatial attention aggregation properties of graph attention networks are leveraged to fuse multimodal features, thereby fully exploiting contextual semantic information. Experimental results on real-world rail surface defect datasets from domestic railway lines demonstrate that the proposed method achieves 95% diagnostic accuracy, confirming its effectiveness in rail surface defect detection. Full article

This paper presents a novel wayside rail monitoring system for real-time detection and quantification of rail surface cracks with sub-millimeter precision. The core innovation lies in mounting piezoelectric transducers on the web of the rail—an unconventional and practical location that avoids interference with wheel passages while enabling continuous monitoring in real-world conditions. Moreover, to directly quantify crack depth, a customized signal processing pipeline is developed, employing surface acoustic waves (SAWs) and incorporating a parallel reference transducer pair mounted on an undamaged rail section for calibration. This auxiliary pair provides a real-time calibration baseline, improving measurement robustness and accuracy. The method is experimentally validated on rail samples and verified through metallographic analysis. This approach enables condition-based maintenance by improving detection accuracy and offers the potential to reduce operational costs and enhance railway safety. Full article

With the widespread application of high-speed and heavy-load railways, the real-time detection of track surface defects has become increasingly crucial. To address the challenges in rail defect detection, this study proposes the PerMSCA-YOLO model, which aims to overcome the limitations of traditional object detection models in multi-scale, small target, and complex background scenarios. By incorporating the lightweight FasterNet backbone network, a multi-scale convolutional attention module, and perceptual loss, the proposed model significantly enhances the detection accuracy and robustness of track defects. Experimental results show that PerMSCA-YOLO achieves an mAP@0.5 of 0.856, an F1-score of 0.79, and an inference frame rate of 142 FPS, demonstrating superior detection accuracy and real-time performance compared to other mainstream models like YOLOv8n. Furthermore, the model exhibits strong adaptability and efficiency when dealing with complex track defects, such as microcracks and corrosion patches, indicating its broad practical application potential. The innovative contribution of this research lies in its effective strategy for improving detection accuracy and real-time performance through multi-scale feature fusion and deep semantic alignment mechanisms, providing a solution that balances both precision and efficiency for defect detection in complex track environments, with substantial engineering application potential. Full article

This work presents the construction of a transformation for the identification of surface defects on rails, starting with the selection of elements from the matrix and the creation of different matrices. It further elaborates on the recursive formulation of the transformation and demonstrates that, regardless of the elements’ uniqueness, the sum of the transformed matrix remains equal to the sum of the original matrix. This study also addresses the handling of matrices with repeated elements and proves that the G-transformation preserves information, ensuring the integrity of data without any loss or redundancy. Full article

The sliding suction robots designed for wall-climbing functions could have accuracy defects due to suction cup sealing, friction interference, and surface adaptability. Hence, this work develops a biomimetic, flexible, sliding suction cup suitable for crawling on curved surfaces. Inspired by the hypostomus plecostomus’s mouth, we designed a biomimetic low-contact force flow channel structure and a matrix of friction-reducing protrusions along the lip edge of the sliding suction cup. This design reduces frictional resistance on the sliding interface and the flexible nature of the suction cup, allowing it to be used on curved or vertical surfaces of different materials. Several simulation-based optimization analyses and experimental tests are conducted on the biomimetic low-contact force flow channel structure, and various structural design principles are explored for achieving high adhesion and low-contact force. Additionally, a friction reduction model for the matrix structure is designed to verify the effects of parameters such as load, protrusion size, and quantity on the friction coefficient of the matrix structure surface through friction tests. The sliding suction cup prototype presents an average crawling speed of about 0.4 m/s on a horizontal plane and 0.7 m/s for crawling on vertical walls and the inner surface of a cylindrical rail. Full article

The electromagnetic railgun, a novel kinetic energy weapon, has found utility in military operations due to its enhanced safety features and superior precision. This study investigates the enhancement of wear resistance in CuCrZr rails through the plasma cladding of CuCrZr-CeO₂ coatings with a varying Cerium dioxide (CeO₂) content. To enhance the wear resistance of the CuCrZr track, plasma cladding of CuCrZr-CeO₂ coatings with varying CeO₂ content was investigated. The impact of CeO₂ content (0%, 0.1%, 0.15%, 0.2%, 0.25%, 0.3%) on the microstructure, phase composition, and mechanical properties of the CuCrZr coating was assessed using scanning electron microscopy (SEM), X-ray diffraction (XRD), EDS (Energy Dispersive Spectrometer) surface scanning, friction and wear tests, and hardness analysis. The findings indicate that a CeO₂ content of 0.15% leads to a transition in the coating’s microstructure from columnar to equiaxed crystals, with the densest grain structure. Beyond 0.15% CeO₂, pore defects in the coating increase, compromising mechanical properties. The coating containing 0.15% CeO₂ exhibits optimal performance, with a hardness of 75.3, representing a 5.31% increase compared to CeO₂-free CuCrZr coatings. Under a 10 N load, the friction coefficient decreases by approximately 17.9% to about 0.64. Moreover, the minimum wear mass is reduced by 44.7% to 3.87 mg. The aforementioned research findings hold immense importance in extending the lifespan of the electromagnetic railgun and improving its operational efficiency. Full article

Detecting track geometry and rail surface defects using on-board vehicle monitoring systems is a key issue for rail infrastructure managers to increase availability and reliability while reducing the costs associated with monitoring and maintenance. Rail corrugation is one of the most common rail surface defects which grows in almost all metro, conventional and high-speed lines. This paper focuses on the development of a methodology to detect rail corrugation using axle box acceleration measurements acquired on an in-service high-speed vehicle. The main purpose of the proposed methodology is to distinguish the effect of rail corrugation on the accelerations from the other excitations that can be observed in the same wavelength range. For this purpose, the accelerations are analysed by calculating the fast Fourier transform and the spectrogram. Based on the characteristics of each excitation, the effects of modes of vibration, resonances, bridges, switches, and wheel defects are identified. From the remaining effects, which have congruent characteristics, a hypothesis of rail corrugation is formulated. The hypothesis is consolidated with multibody dynamics simulations and by comparing the corrugation indicators provided by the railway infrastructure company. Full article

Rail surface defect detection is vital for railway safety. Traditional methods falter with varying defect sizes and complex backgrounds, while two-stage deep learning models, though accurate, lack real-time capabilities. To overcome these challenges, we propose an enhanced one-stage detection model based on CenterNet. We replace ResNet with ResNeXt and implement a multi-branch structure for better low-level feature extraction. Additionally, we integrate SKNet attention mechanism with the C2f structure from YOLOv8, improving the model’s focus on critical image regions and enhancing the detection of minor defects. We also introduce an elliptical Gaussian kernel for size regression loss, better representing the aspect ratio of rail defects. This approach enhances detection accuracy and speeds up training. Our model achieves a mean accuracy (mAP) of 0.952 on the rail defects dataset, outperforming other models with a 6.6% improvement over the original and a 35.5% increase in training speed. These results demonstrate the efficiency and reliability of our method for rail defect detection. Full article

Railways play a pivotal role in national economic development, freight transportation, national defense, and regional connectivity. The detection of rolling contact fatigue (RCF) defects in rail tracks is essential for railway safety and maintenance. Due to its efficiency and non-contact capability in detecting surface and near-surface defects, Eddy Current Pulsed Thermography (ECPT) has garnered significant attention from researchers. However, detecting multi-layer RCF defects remains a challenge. This paper introduces a sweeping Eddy Current Pulsed Thermal-Tomography system (ECPTT) to detect multi-layer RCF defects effectively. This system utilizes varying excitation frequencies to heat defects, altering skin depth and facilitating feature extraction to distinguish multi-layer RCF defects. Skewness and thermographic signal reconstruction (TSR) values are employed as features in the experiments. These features are qualitatively analyzed to differentiate the layers and depths of multi-layer RCF defects. Additionally, five different coils were compared and analyzed quantitatively. The results indicate that the ECPTT system can effectively detect and distinguish multi-layer RCF defects, thereby providing more detailed defect information and enhancing railway safety and maintenance efficiency. Full article

Rail surface defects typically serve as early indicators of railway malfunctions, which may compromise the quality and corrosion resistance of rails, thereby endangering the safe operation of trains. The timely detection of defects is essential to ensure the safe operation of railways. To improve the classification accuracy of rail surface defect detection, this paper proposes a rail surface defects detection algorithm based on MobileNet-YOLOv7. By integrating lightweight deep learning algorithms into the engineering application of rail surface defect detection, a MobileNetV3 lightweight network is used as the backbone network for YOLOv7 to enhance both speed and accuracy in complex defect extraction. Subsequently, the efficient intersection over union (EIOU) loss function is utilized as the positional loss function to bolster system resilience. Finally, the k-means++ clustering algorithm is applied to obtain new anchor boxes. The experimental results demonstrate the effectiveness of the proposed method, achieving superior detection accuracy compared with traditional algorithms. Full article

A wheel flat is a typical wheel defect that significantly impacts the wheel–rail system, posing substantial challenges to vehicle operation safety. In the existing literature, the wheel flat plane model does not account for the contribution of the width direction to the impact response and thus cannot accurately reveal the wheel–rail contact state with a flat. This paper systematically proposes a three-dimensional analytical model that considers multiple worn stages and constructs a spatial complex surface reconstruction model for flats based on NURBS technology. A vehicle–track coupled dynamics model, considering the geometry of the flat, is established to investigate the effects of flat geometry on the wheel–rail impact response and contact relationship in detail. The results show that in the subcritical regime, the wear degree of the flat predominantly affects the impact force, while in the transcritical regime, both the wear degree and velocity together determine the magnitude of the wheel–rail impact force. As the wear degree increases, the moment of wheel lateral jump occurs earlier. The spatial modeling method for flats proposed in this paper offers a novel technical approach for accurately simulating the dynamic behavior of wheel–rail contact when a flat is present. Full article