To guarantee the safety of railway operations,the swift detection of rail surface defects becomes imperative.Traditional methods of manual inspection and conventional nondestructive testing prove inefficient,especiall...To guarantee the safety of railway operations,the swift detection of rail surface defects becomes imperative.Traditional methods of manual inspection and conventional nondestructive testing prove inefficient,especially when scaling to extensive railway networks.Moreover,the unpredictable and intricate nature of defect edge shapes further complicates detection efforts.Addressing these challenges,this paper introduces an enhanced Unified Perceptual Parsing for Scene Understanding Network(UPerNet)tailored for rail surface defect detection.Notably,the Swin Transformer Tiny version(Swin-T)network,underpinned by the Transformer architecture,is employed for adept feature extraction.This approach capitalizes on the global information present in the image and sidesteps the issue of inductive preference.The model’s efficiency is further amplified by the windowbased self-attention,which minimizes the model’s parameter count.We implement the cross-GPU synchronized batch normalization(SyncBN)for gradient optimization and integrate the Lovász-hinge loss function to leverage pixel dependency relationships.Experimental evaluations underscore the efficacy of our improved UPerNet,with results demonstrating Pixel Accuracy(PA)scores of 91.39%and 93.35%,Intersection over Union(IoU)values of 83.69%and 87.58%,Dice Coefficients of 91.12%and 93.38%,and Precision metrics of 90.85%and 93.41%across two distinct datasets.An increment in detection accuracy was discernible.For further practical applicability,we deploy semantic segmentation of rail surface defects,leveraging connected component processing techniques to distinguish varied defects within the same frame.By computing the actual defect length and area,our deep learning methodology presents results that offer intuitive insights for railway maintenance professionals.展开更多
A more effective and accurate improved Sobel algorithm has been developed to detect surface defects on heavy rails. The proposed method can make up for the mere sensitivity to X and Y directions of the Sobel algorithm...A more effective and accurate improved Sobel algorithm has been developed to detect surface defects on heavy rails. The proposed method can make up for the mere sensitivity to X and Y directions of the Sobel algorithm by adding six templates at different directions. Meanwhile, an experimental platform for detecting surface defects consisting of the bed-jig, image-forming system with CCD cameras and light sources, parallel computer system and cable system has been constructed. The detection results of the backfin defects show that the improved Sobel algorithm can achieve an accurate and efficient positioning with decreasing interference noises to the defect edge. It can also extract more precise features and characteristic parameters of the backfin defect. Furthermore, the BP neural network adopted for defects classification with the inputting characteristic parameters of improved Sobel algorithm can obtain the optimal training precision of 0.0095827 with 106 iterative steps and time of 3 s less than Sobel algorithm with 146 steps and 5 s. Finally, an enhanced identification rate of 10% for the defects is also confirmed after the Sobel algorithm is improved.展开更多
A novel electromagnetic tomography(EMT)system for defect detection of high-speed rail wheel is proposed,which differs from traditional electromagnetic tomography systems in its spatial arrangements of coils.A U-shaped...A novel electromagnetic tomography(EMT)system for defect detection of high-speed rail wheel is proposed,which differs from traditional electromagnetic tomography systems in its spatial arrangements of coils.A U-shaped sensor array was designed,and then a simulation model was built with the low frequency electromagnetic simulation software.Three different algorithms were applied to perform image reconstruction,therefore the defects can be detected from the reconstructed images.Based on the simulation results,an experimental system was built and image reconstruction were performed with the measured data.The reconstructed images obtained both from numerical simulation and experimental system indicated the locations of the defects of the wheel,which verified the feasibility of the EMT system and revealed its good application prospect in the future.展开更多
On-line rail milling technologies have been applied in rail maintenance, and are proving to be efficient and environmental friendly. Based on the field data of on-line rail milling, a program for comparing rail transv...On-line rail milling technologies have been applied in rail maintenance, and are proving to be efficient and environmental friendly. Based on the field data of on-line rail milling, a program for comparing rail transverse profiles before and after milling was designed and the root mean square （RMS） amplitude of longitudinal profile was calculated. The application of on-line rail milling technology in removing rail surface defects, re-profiling railhead transverse profiles, smoothing longitudinal profiles and improving welding joint irregularity were analyzed. The results showed that the on-line rail milling technology can remove the surface defects at the rail crown and gauge comer perfectly, re-profile railhead transverse profile with a tolerance of - 1. 0-0.2 ram, improve longitudinal irregularity of rail surface, with the RMS amplitude of irregularity reduced more than 50% and the number of out-of- limited amplitude reduced by 42% - 82% in all wavelength ranges. The improvement of welding joint irregularity depends on the amount of metal removal determined by the milling equipment and the primal amplitude.展开更多
针对钢轨表面缺陷检测效率较低及抗干扰能力较差的问题,提出一种基于改进YOLOv5的钢轨表面缺陷检测算法.首先,采用图像增强操作对采集到的钢轨表面图像进行预处理,减轻高光、异物等噪声对检测效果的影响.其次,将多头自注意力层嵌入YOLOv...针对钢轨表面缺陷检测效率较低及抗干扰能力较差的问题,提出一种基于改进YOLOv5的钢轨表面缺陷检测算法.首先,采用图像增强操作对采集到的钢轨表面图像进行预处理,减轻高光、异物等噪声对检测效果的影响.其次,将多头自注意力层嵌入YOLOv5骨干网络末端,并为缺陷特征引入全局依赖关系,提升模型对密集缺陷的检测效果.最后,构建跨层加权级联结构,将浅层信息融入到深层网络中,使网络对缺陷边界的回归更为精准.实验结果表明:本文的钢轨表面缺陷检测算法对裂纹、剥落、磨损3类表面缺陷检测的平均精度均值达到98.2%,每秒帧数(Frames Per Second,FPS)达到77帧/s,能够在不同的环境条件中实现对缺陷的精准检测,比其他某些同类算法拥有更高的鲁棒性、准确性和实时性.展开更多
针对现有基于深度学习的钢轨表面缺陷检测方法在嵌入式检测系统上兼容性较差、计算资源占用高以及检测速度慢的问题,提出了一种基于改进YOLOX的轻量级钢轨表面缺陷检测算法。模型中主干特征层以MobileNetv3单元为基础,在保留其网络轻量...针对现有基于深度学习的钢轨表面缺陷检测方法在嵌入式检测系统上兼容性较差、计算资源占用高以及检测速度慢的问题,提出了一种基于改进YOLOX的轻量级钢轨表面缺陷检测算法。模型中主干特征层以MobileNetv3单元为基础,在保留其网络轻量化的同时进行局部优化,改进了浅层网络的激活函数,嵌入了SE(Squeeze and Excitation)注意力机制;在加强特征层优化了尾部的冗余卷积。通过与几种代表性算法进行对比试验,验证该算法的性能。结果表明:本文提出的改进算法在模型参数量仅为1.10×106的情况下,检出率和准确率分别达到了92.17%和90.92%,每秒传输帧数(Frame Per Second,FPS)为115.07,模型大小仅为原模型的1/5。该算法在保证较高检测精度的同时大大降低了模型参数量,并提升了检测速度,更适合部署于算力有限的嵌入式轨道检测系统,可为钢轨缺陷高效检测提供有效手段。展开更多
基金supported in part by the National Natural Science Foundation of China(Grant No.62066024)Gansu Province Higher Education Industry Support Plan(2021CYZC34)Lanzhou Talent Innovation and Entrepreneurship Project(2021-RC-27,2021-RC-45).
文摘To guarantee the safety of railway operations,the swift detection of rail surface defects becomes imperative.Traditional methods of manual inspection and conventional nondestructive testing prove inefficient,especially when scaling to extensive railway networks.Moreover,the unpredictable and intricate nature of defect edge shapes further complicates detection efforts.Addressing these challenges,this paper introduces an enhanced Unified Perceptual Parsing for Scene Understanding Network(UPerNet)tailored for rail surface defect detection.Notably,the Swin Transformer Tiny version(Swin-T)network,underpinned by the Transformer architecture,is employed for adept feature extraction.This approach capitalizes on the global information present in the image and sidesteps the issue of inductive preference.The model’s efficiency is further amplified by the windowbased self-attention,which minimizes the model’s parameter count.We implement the cross-GPU synchronized batch normalization(SyncBN)for gradient optimization and integrate the Lovász-hinge loss function to leverage pixel dependency relationships.Experimental evaluations underscore the efficacy of our improved UPerNet,with results demonstrating Pixel Accuracy(PA)scores of 91.39%and 93.35%,Intersection over Union(IoU)values of 83.69%and 87.58%,Dice Coefficients of 91.12%and 93.38%,and Precision metrics of 90.85%and 93.41%across two distinct datasets.An increment in detection accuracy was discernible.For further practical applicability,we deploy semantic segmentation of rail surface defects,leveraging connected component processing techniques to distinguish varied defects within the same frame.By computing the actual defect length and area,our deep learning methodology presents results that offer intuitive insights for railway maintenance professionals.
基金Project(51174151)supported by the National Natural Science Foundation of ChinaProject(2010Z19003)supported by the Major Scientific Research Program of Hubei Provincial Department of Education,ChinaProject(2010CDB03403)supported by the Natural Science Foundation of Science and Technology Department of Hubei Province,China
文摘A more effective and accurate improved Sobel algorithm has been developed to detect surface defects on heavy rails. The proposed method can make up for the mere sensitivity to X and Y directions of the Sobel algorithm by adding six templates at different directions. Meanwhile, an experimental platform for detecting surface defects consisting of the bed-jig, image-forming system with CCD cameras and light sources, parallel computer system and cable system has been constructed. The detection results of the backfin defects show that the improved Sobel algorithm can achieve an accurate and efficient positioning with decreasing interference noises to the defect edge. It can also extract more precise features and characteristic parameters of the backfin defect. Furthermore, the BP neural network adopted for defects classification with the inputting characteristic parameters of improved Sobel algorithm can obtain the optimal training precision of 0.0095827 with 106 iterative steps and time of 3 s less than Sobel algorithm with 146 steps and 5 s. Finally, an enhanced identification rate of 10% for the defects is also confirmed after the Sobel algorithm is improved.
基金Supported by the National Natural Science Foundation of China(61771041)。
文摘A novel electromagnetic tomography(EMT)system for defect detection of high-speed rail wheel is proposed,which differs from traditional electromagnetic tomography systems in its spatial arrangements of coils.A U-shaped sensor array was designed,and then a simulation model was built with the low frequency electromagnetic simulation software.Three different algorithms were applied to perform image reconstruction,therefore the defects can be detected from the reconstructed images.Based on the simulation results,an experimental system was built and image reconstruction were performed with the measured data.The reconstructed images obtained both from numerical simulation and experimental system indicated the locations of the defects of the wheel,which verified the feasibility of the EMT system and revealed its good application prospect in the future.
基金The National Natural Science Foundation of China(No.50908179)Specialized Research Fund for the Doctoral Program of Higher Education(No.200802471003)Program for Young Excellent Talents in Tongji University(No.2008KJ026)
文摘On-line rail milling technologies have been applied in rail maintenance, and are proving to be efficient and environmental friendly. Based on the field data of on-line rail milling, a program for comparing rail transverse profiles before and after milling was designed and the root mean square （RMS） amplitude of longitudinal profile was calculated. The application of on-line rail milling technology in removing rail surface defects, re-profiling railhead transverse profiles, smoothing longitudinal profiles and improving welding joint irregularity were analyzed. The results showed that the on-line rail milling technology can remove the surface defects at the rail crown and gauge comer perfectly, re-profile railhead transverse profile with a tolerance of - 1. 0-0.2 ram, improve longitudinal irregularity of rail surface, with the RMS amplitude of irregularity reduced more than 50% and the number of out-of- limited amplitude reduced by 42% - 82% in all wavelength ranges. The improvement of welding joint irregularity depends on the amount of metal removal determined by the milling equipment and the primal amplitude.
文摘针对钢轨表面缺陷检测效率较低及抗干扰能力较差的问题,提出一种基于改进YOLOv5的钢轨表面缺陷检测算法.首先,采用图像增强操作对采集到的钢轨表面图像进行预处理,减轻高光、异物等噪声对检测效果的影响.其次,将多头自注意力层嵌入YOLOv5骨干网络末端,并为缺陷特征引入全局依赖关系,提升模型对密集缺陷的检测效果.最后,构建跨层加权级联结构,将浅层信息融入到深层网络中,使网络对缺陷边界的回归更为精准.实验结果表明:本文的钢轨表面缺陷检测算法对裂纹、剥落、磨损3类表面缺陷检测的平均精度均值达到98.2%,每秒帧数(Frames Per Second,FPS)达到77帧/s,能够在不同的环境条件中实现对缺陷的精准检测,比其他某些同类算法拥有更高的鲁棒性、准确性和实时性.
文摘针对现有基于深度学习的钢轨表面缺陷检测方法在嵌入式检测系统上兼容性较差、计算资源占用高以及检测速度慢的问题,提出了一种基于改进YOLOX的轻量级钢轨表面缺陷检测算法。模型中主干特征层以MobileNetv3单元为基础,在保留其网络轻量化的同时进行局部优化,改进了浅层网络的激活函数,嵌入了SE(Squeeze and Excitation)注意力机制;在加强特征层优化了尾部的冗余卷积。通过与几种代表性算法进行对比试验,验证该算法的性能。结果表明:本文提出的改进算法在模型参数量仅为1.10×106的情况下,检出率和准确率分别达到了92.17%和90.92%,每秒传输帧数(Frame Per Second,FPS)为115.07,模型大小仅为原模型的1/5。该算法在保证较高检测精度的同时大大降低了模型参数量,并提升了检测速度,更适合部署于算力有限的嵌入式轨道检测系统,可为钢轨缺陷高效检测提供有效手段。
