结合空间域与频率域处理的钢轨表面缺陷检测研究

doi:10.3778/j.issn.1002-8331.2405-0236

摘要/Abstract

摘要： 针对钢轨表面缺陷检测任务存在因相机抖动造成的图像模糊失真而影响检测效果的问题，提出了一种结合空间域与频率域处理的图像增强算法与改进后YOLOv8s模型的钢轨表面缺陷检测算法，旨在通过提高图像质量来提高检测网络的检测精度。通过图像恢复网络对图像去模糊复原，使用深度学习缩减检测区域，提出了空间域的全局与局部信息融合的自适应伽马变换和基于拉普拉斯金字塔改进的频率域滤波算法来提升图像亮度、对比度和细节信息。借助YOLOv8s算法对增强后的图像进行训练，并通过改进特征融合网络、引入注意力模块和设计基于参数共享的检测头来降低网络参数，提升网络检测精度。实验结果表明，相比于未使用图像增强算法，该算法增强后数据集的检测精度有了更好的效果；采用增强后的数据集进行训练，相比于原网络，改进后的检测网络对所有目标、大目标和小目标的平均精度均值都有明显提升。

关键词: 钢轨表面缺陷, 自适应伽马变换, 频率域滤波, YOLOv8s, 参数共享

Abstract: Aiming at the problem of image blur and distortion caused by camera shake, which affects the detection performance in the task of rail surface defect detection, this paper proposes an image enhancement algorithm that combines spatial domain and frequency domain processing. Additionally, an improved YOLOv8s model is presented for rail surface defect detection. The purpose is to enhance image quality to improve the detection accuracy of the detection network. The images are deblurred and restored using image restoration network, with deep learning utilized to reduce the detection region. An adaptive gamma transformation, which integrates global and local spatial information, along with a frequency domain filtering algorithm improved via Laplacian pyramids, is proposed to enhance image brightness, contrast, and detail information. Furthermore, the enhanced images are trained using the YOLOv8s algorithm, with enhancements in the feature fusion network, the integration of attention modules, and the design of a detection head based on parameter sharing to reduce network parameters and increase detection precision. Experimental results demonstrate that compared to datasets not using the image enhancement algorithm, the enhanced dataset achieves better detection accuracy; when trained on the enhanced dataset, the improved detection network shows a significant improvement in mean average precision across all targets, large targets, and small targets.

Key words: rail surface defects, adaptive gamma transformation, frequency domain filtering, YOLOv8s, parameter sharing

冀萌浩, 王新. 结合空间域与频率域处理的钢轨表面缺陷检测研究[J]. 计算机工程与应用, 2025, 61(19): 348-358.

JI Menghao, WANG Xin. Research on Detection of Rail Surface Defects by Combining Spatial Domain and Frequency Domain Processing Techniques[J]. Computer Engineering and Applications, 2025, 61(19): 348-358.

参考文献

[1] 赵宏伟, 郑嘉俊, 赵鑫欣, 等. 基于双模态深度学习的钢轨表面缺陷检测方法[J]. 计算机工程与应用, 2023, 59(7): 285-293.
ZHAO H W, ZHENG J J, ZHAO X X, et al. Rail surface defect method based on bimodal-modal deep learning[J]. Computer Engineering and Applications, 2023, 59(7): 285-293.
[2] 杨泽青, 张明轩, 陈英姝, 等. 基于机器视觉的表面缺陷检测方法研究进展[J]. 现代制造工程, 2023(4): 143-156.
YANG Z Q, ZHANG M X, CHEN Y S, et al. Review of surface defect detection methods based on machine vision[J]. Modern Manufacturing Engineering, 2023(4): 143-156.
[3] ZHANG A Z, HU X Y, JIN M Y, et al. Multi-target defect detection of railway track based on image processing[C]//Proceedings of the 2020 Chinese Control and Decision Conference. Piscataway: IEEE, 2020: 3377-3382.
[4] 罗晖, 徐广隆. 基于图像增强与深度学习的钢轨表面缺陷检测[J]. 铁道科学与工程学报, 2021, 18(3): 623-629.
LUO H, XU G L. Rail surface defect detection based on image enhancement and deep learning[J]. Journal of Railway Science and Engineering, 2021, 18(3): 623-629.
[5] 岳彪, 闵永智, 马宏锋, 等. 钢轨表面缺陷检测系统中图像增强预处理方法研究[J]. 铁道科学与工程学报, 2018, 15(12): 3248-3256.
YUE B, MIN Y Z, MA H F, et al. Research on image enhancement preprocessing method in rail surface defects detection system[J]. Journal of Railway Science and Engineering, 2018, 15(12): 3248-3256.
[6] 杨超. 时频域结合的数字图像增强技术的研究[D]. 唐山: 华北理工大学, 2019.
YANG C. Research on digital image enhancement technology based on spatial and frequency domain[D]. Tangshan: North China University of Science and Technology, 2019.
[7] LI J I, LUO P, HE S X, et al. Application of FTP in rail profile and surface flaw detection based on orthogonal two-frequency composite grating[C]//Proceedings of the 2017 Far East NDT New Technology & Application Forum. Piscataway: IEEE, 2017: 318-322.
[8] 王正家, 昝傲, 谷峰. 基于Gabor滤波的软包电池表面缺陷检测[J]. 组合机床与自动化加工技术, 2023(11): 146-149.
WANG Z J, ZAN A, GU F. Surface defect detection of pouch cells based on Gabor filtering[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2023(11): 146-149.
[9] 杜少聪, 张红钢, 王小敏. 基于改进YOLOv5的钢轨表面缺陷检测[J]. 北京交通大学学报, 2023, 47(2): 129-136.
DU S C, ZHANG H G, WANG X M. Rail surface defect detection based on improved YOLOv5[J]. Journal of Beijing Jiaotong University, 2023, 47(2): 129-136.
[10] 刘俊博, 杜馨瑜, 王胜春, 等. 基于少样本学习的钢轨表面缺陷检测方法[J]. 铁道学报, 2022, 44(7): 72-79.
LIU J B, DU X Y, WANG S C, et al. Rail surface defect detection method based on few-shot learning[J]. Journal of the China Railway Society, 2022, 44(7): 72-79.
[11] ANWAR N, SHEN Z, WEI Q L, et al. YOLOv4 based deep learning algorithm for defects detection and classification of rail surfaces[C]//Proceedings of the 2021 IEEE International Intelligent Transportation Systems Conference. Piscataway: IEEE, 2021: 1616-1620.
[12] 邢艳, 陈晓璐, 徐启奥, 等. 融合空间域和频率域信息的图像去模糊[J]. 计算机系统应用, 2024, 33(2): 1-12.
XING Y, CHEN X L, XU Q A, et al. Image deblurring by fusing information of spatial and frequency domains[J]. Computer Systems and Applications, 2024, 33(2): 1-12.
[13] ZAMIR S W, ARORA A, KHAN S, et al. Multi-stage progressive image restoration[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2021: 14816-14826.
[14] 李超超, 张远东, 王昌龙. 圆形植物识别定位的HSV模型研究[J]. 计算机工程与应用, 2020, 56(6): 220-223.
LI C C, ZHANG Y D, WANG C L. Research on HSV model for identifying and locating round plants[J]. Computer Engineering and Applications, 2020, 56(6): 220-223.
[15] 张弢, 蒋东东, 田喆文, 等. 基于ACE与YOLOv5的电力遥感图像检测算法[J]. 计算机仿真, 2024, 41(1): 277-283.
ZHANG T, JIANG D D, TIAN Z W, et al. Electric power remote sensing image detection algorithm based on automatic color balance and YOLOv5[J]. Computer Simulation, 2024, 41(1): 277-283.
[16] REDDY G R, SRINIVAS A, GIRIJA S P, et al. Enhancement of images using optimized gamma correction with weighted distribution via differential evolution algorithm[C]//Proceedings of the 2022 First International Conference on Electrical, Electronics, Information and Communication Technologies. Piscataway: IEEE, 2022: 1-5.
[17] HUANG L H, CAO G, YU L F. Efficient contrast enhancement with truncated adaptive gamma correction[C]//Proceedings of the 2016 9th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics. Piscataway: IEEE, 2016: 189-194.
[18] SAHNOUN M, KALLEL F, DAMMAK M, et al. A comparative study of MRI contrast enhancement techniques based on traditional gamma correction and adaptive gamma correction: case of multiple sclerosis pathology[C]//Proceedings of the 2018 4th International Conference on Advanced Technologies for Signal and Image Processing. Piscataway: IEEE, 2018: 1-7.
[19] 王大红, 胡茂林. 巴特沃斯非线性混合滤波器图像滤波方法设计[J]. 计算机工程与应用, 2010, 46(21): 195-198.
WANG D H, HU M L. Image processing method using mixed non-linear Butterworth filter[J]. Computer Engineering and Applications, 2010, 46(21): 195-198.
[20] BURT P J, ADELSON E H. The Laplacian pyramid as a compact image code[J]. IEEE Transactions on Communications, 1983, 31(4): 532-540.
[21] XIAO J S, ZHAO T, YAO Y T, et al. Context augmentation and feature refinement network for tiny object detection [C]//Proceedings of the 10th International Conference on Learning Representations, 2022.
[22] XIA Z F, PAN X R, SONG S J, et al. Vision transformer with deformable attention[C]//Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2022: 4784-4793.
[23] CHEN J R, KAO S H, HE H, et al. Run, don’t walk: chasing higher FLOPS for faster neural networks[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2023: 12021-12031.
[24] WANG Z, BOVIK A C, SHEIKH H R, et al. Image quality assessment: from error visibility to structural similarity[J]. IEEE Transactions on Image Processing, 2004, 13(4): 600-612.
[25] ZHANG R, ISOLA P, EFROS A A, et al. The unreasonable effectiveness of deep features as a perceptual metric[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 586-595.
[26] 史涛, 崔杰, 李松. 优化改进YOLOv8实现实时无人机车辆检测的算法[J]. 计算机工程与应用, 2024, 60(9): 79-89.
SHI T, CUI J, LI S. Algorithm for real-time vehicle detection from UAVs based on optimizing and improving YOLOv8[J]. Computer Engineering and Applications, 2024, 60(9): 79-89.
[27] MITTAL A, MOORTHY A K, BOVIK A C. Blind/referenceless image spatial quality evaluator[C]//Proceedings of the 2011 Conference Record of the 45th Asilomar Conference on Signals, Systems and Computers. Piscataway: IEEE, 2011: 723-727.
[28] MITTAL A, SOUNDARARAJAN R, BOVIK A C. Making a “completely blind” image quality analyzer[J]. IEEE Signal Processing Letters, 2013, 20(3): 209-212.
[29] ROZUMNYI D, OSWALD M R, FERRARI V, et al. DeFMO: deblurring and shape recovery of fast moving objects[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2021: 3455-3464.
[30] TAO X, GAO H Y, SHEN X Y, et al. Scale-recurrent network for deep image deblurring[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 8174-8182.
[31] NAH S, KIM T H, LEE K M. Deep multi-scale convolutional neural network for dynamic scene deblurring[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 257-265.
[32] ZHANG J M, LV Y R, TAO J J, et al. A robust real-time anchor-free traffic sign detector with one-level feature[J]. IEEE Transactions on Emerging Topics in Computational Intelligence, 2024, 8(2): 1437-1451.
[33] YANG W H, YUAN Y, REN W Q, et al. Advancing image understanding in poor visibility environments: a collective benchmark study[J]. IEEE Transactions on Image Processing, 2020, 29: 5737-5752.
[34] BAO Y Q, SONG K C, LIU J, et al. Triplet-graph reasoning network for few-shot metal generic surface defect segmentation[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-11.
[35] ABDULLAH-AL-WADUD M, KABIR M H, DEWAN M A A, et al. A dynamic histogram equalization for image contrast enhancement[C]//Proceedings of the 2007 Digest of Technical Papers International Conference on Consumer Electronics. Piscataway: IEEE, 2007: 1-2.
[36] LEE C, LEE C, KIM C S. Contrast enhancement based on layered difference representation of 2D histograms[J]. IEEE Transactions on Image Processing, 2013, 22(12): 5372-5384.
[37] CELIK T, TJAHJADI T. Contextual and variational contrast enhancement[J]. IEEE Transactions on Image Processing, 2011, 20(12): 3431-3441.
[38] YING Z Q, LI G, GAO W. A bio-inspired multi-exposure fusion framework for low-light image enhancement[J]. arXiv:1711.00591, 2017.
[39] LIU R S, MA L, ZHANG J A, et al. Retinex-inspired unrolling with cooperative prior architecture search for low-light image enhancement[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2021: 10556-10565.
[40] CAI Y H, BIAN H, LIN J, et al. Retinexformer: one-stage retinex-based transformer for low-light image enhancement[C]//Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2023: 12470-12479.
[41] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proceedings of the 15th European Conference on Computer Vision. Cham: Springer, 2018: 3-19.
[42] HUANG H J, CHEN Z G, ZOU Y, et al. Channel prior convolutional attention for medical image segmentation[J]. Computers in Biology and Medicine, 2024, 178: 108784.