An Improved YOLOv8 Algorithm for Rail Surface Defect Detection
Open accessThe DOI matches its Crossref record; 32 of the 33 references it cites were found in a registry or an index.
Abstract
To tackle the issues raised by detecting small targets and densely occluded targets in railroad track surface defect detection, we present an algorithm for detecting defects on railroad tracks based on the YOLOv8 model. Firstly, we enhance the model’s attention towards small and medium-sized targets by substituting replacing the original convolution with the SPD-Conv building block in the backbone network of YOLOv8n, while preserving the original network structure. Secondly, we integrate the integrating the EMA attention mechanism module into the neck component, allowing the model to leverage information from different layers of features and improve feature representation capabilities. Lastly, we substitute the original C-IOU with the Focal-SIoU loss function in YOLOv8., which adjusts the weights of positive and negative samples to penalize difficult-to-classify samples more heavily. This enhancement improves the model’s capability to accurately recognize challenging samples and ensures that the network allocates greater attention to each target instance, resulting in improved performance and effectiveness of the model. The experimental results reveal notable advancements in precision, recall, and average accuracy attained by our enhanced algorithm. Compared to the original YOLOv8n model, our enhanced algorithm demonstrates remarkable precision, recall, and average accuracy of 93.9%, 93.7%, and 94.1%, respectively. These improvements amount to 3.6%, 5.0%, and 5.7%, respectively. Notably, these enhancements are accomplished while maintaining the dimensions of the model and the parameter count. During the identification of defects on railroad track surfaces, our improved algorithm surpasses other widely used algorithms in terms of performance.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures



















Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The deep learning-based YOLOv8 model is used to detect the object and its applications are diverse including road accident detection [48].”
From Visual Detection of Traffic Incident through Automatic Monitoring of Vehicle Activities (2024) · cites this paper as [48]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Focal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.324 · Cited by 30,581 · cited 1×
“(c)The Focal-Segmentation Intersection over Union (Focal-SIoU) [26], [27] loss function is utilized as a replacement for the original Complete-Intersection over Union (C-IoU) loss function.”
MethodYOLOv4: Optimal Speed and Accuracy of Object Detection
Bochkovskiy, A., Wang, C.-Y., & Liao, H.-Y. M. (2020) · arXiv · doi:10.48550/arxiv.2004.10934 · Cited by 10,307 · cited 1×
“(4) In the data augmentation step during training, an operation of disabling Mosiac enhancement [34] for the last 10 epochs in YOLOX was introduced, resulting in improved accuracy.”
BackgroundDistance-IoU Loss: Faster and Better Learning for Bounding Box Regression
Zheng, Z., Wang, P., Liu, W., Li, J., Ye, R., & Ren, D. (2019) · arXiv · doi:10.48550/arxiv.1911.08287 · Cited by 5,313 · cited 1×
“Additionally, CIoU Loss [33] was also incorporated.”
Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression
Zheng, Z., Wang, P., Liu, W., Li, J., Ye, R., & Ren, D. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i07.6999 · Cited by 5,288
YOLOX: Exceeding YOLO Series in 2021
Ge, Z., Liu, S., Wang, F., Li, Z., & Sun, J. (2021) · arXiv · doi:10.48550/arxiv.2107.08430 · Cited by 2,965 · cited 1×
“(2)TheHead section has been considerably changed from YOLOv5 to the prevailing decoupled header structure in use [29], which segregates the classification and detection headers, as well as from Anchor-Based to Anchor-Free [30].”
MethodA Review of Yolo Algorithm Developments
Jiang, P., Ergu, D., Liu, F., Cai, Y., & Ma, B. (2022) · Procedia Computer Science · doi:10.1016/j.procs.2022.01.135 · Cited by 2,783 · cited 1×
“The model series known as You Only Look Once (YOLO) [8], as a widely utilized target detection framework, has been extensively employed in detecting railroad track defects, yielding commendable accuracy and detection outcomes.”
YOLACT: Real-Time Instance Segmentation
Bolya, D., Zhou, C., Xiao, F., & Lee, Y. J. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00925 · Cited by 2,536 · cited 1×
“Additionally, this algorithm has developed an instance segmentation model using the You Only Look At Coefficients (YOLACT) [28] architecture.”
MethodZhang, S., Chi, C., Yao, Y., Lei, Z., & Li, S. Z. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00978 · Cited by 2,354 · cited 1×
“(2)TheHead section has been considerably changed from YOLOv5 to the prevailing decoupled header structure in use [29], which segregates the classification and detection headers, as well as from Anchor-Based to Anchor-Free [30].”
MethodU-Net and Its Variants for Medical Image Segmentation: A Review of Theory and Applications
Siddique, N., Paheding, S., Elkin, C. P., & Devabhaktuni, V. (2021) · IEEE Access · doi:10.1109/access.2021.3086020 · Cited by 1,960 · cited 1×
“Noteworthy examples encompass the adoption of models like U-Net [10] and Mask R-CNN [11] for localizing and segmenting defective regions on the track.”
MethodEfficient Multi-Scale Attention Module with Cross-Spatial Learning
Ouyang, D., He, S., Zhang, G., Luo, M., Guo, H., Zhan, J., & Huang, Z. (2023) · ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp49357.2023.10096516 · Cited by 1,788
Research works printed but not linked: 7
- J.W. Huo, Z. Liu, Y.D. Wang, et al. Planar electromagnetic tomography for rail flaw detection[J]. Chinese Journal of Electrical Engineering,2021,41(15):5351-5361.
- Hu Xiaoqun. Research on the application of nondestructive testing technology in steel structure construction engineering[J]. Real Estate World,2023(19):142-144.
- Xu X, Zhao M, Shi P, et al. Crack detection and comparison study based on faster R-CNN and mask R-CNN[J]. Sensors, 2022, 22(3): 1215.
- Mohan K K, Prasad C R, Kishore P V V. Yolo v2 with bifold skip: a deep learning model for video based real time train bogie part identification and defect detection[J]. J Eng Sci Technol, 2021, 16(3): 2166-2190.
- Cao Hongpeng. Development of visual inspection system for surface defects of cold heavy rail based on deep learning [D]. Zhejiang University,2021.
- SUSTECH 2023. track defect detection Dataset [M]. Roboflow.
- FENG 2023. clipping Dataset
Cited by
ALF-YOLO: Enhanced YOLOv8 based on multiscale attention feature fusion for ship detection
Wang, S., Li, Y., & Qiao, S. (2024) · Ocean Engineering · doi:10.1016/j.oceaneng.2024.118233 · Cited by 114
Lightweight rail surface defect detection algorithm based on an improved YOLOv8
CanYang, X., Yingying, L., Yongqiang, L., Tian, R., & Guo, T. (2024) · Measurement · doi:10.1016/j.measurement.2024.115922 · Cited by 66
A wall climbing robot based on machine vision for automatic welding seam inspection
Du, Y., Liu, M., Wang, J., Liu, X., Wang, K., Liu, Z., Dong, Q., Yao, J., Lu, D., & Su, Y. (2024) · Ocean Engineering · doi:10.1016/j.oceaneng.2024.118825 · Cited by 29
Visual Detection of Traffic Incident through Automatic Monitoring of Vehicle Activities
Karim, A., Raza, M. A., Alharthi, Y. Z., Abbas, G., Othmen, S., Hossain, M. S., Nahar, A., & Mercorelli, P. (2024) · World Electric Vehicle Journal · doi:10.3390/wevj15090382 · Cited by 23
Gupta, C., Gill, N. S., Gulia, P., Duhan, S., Alduaiji, N., Shukla, P. K., & Dwivedi, A. (2025) · BioData Mining · doi:10.1186/s13040-025-00497-y · Cited by 20
“CSP layers divide the feature maps into two parts and use a cross-stage partial approach to boost feature learning while reducing computation [69].”
Fault Detection and Semantic Segmentation on Railway Track Using Deep Fusion Model
Rampriya, R. S., Jenefa, A., Prathiba, S. B., Sabarinathan, Julus, L. J., Selvaraj, A. K., & Rodrigues, J. J. P. C. (2024) · IEEE Access · doi:10.1109/access.2024.3462822 · Cited by 24
Enhancing traffic monitoring with noise-robust distributed acoustic sensing and deep learning
Wang, Z., Zhang, T., Chen, H., Zhang, C., & Shi, B. (2024) · Journal of Applied Geophysics · doi:10.1016/j.jappgeo.2024.105616 · Cited by 22
RSDNet: A New Multiscale Rail Surface Defect Detection Model
Du, J., Zhang, R., Gao, R., Nan, L., & Bao, Y. (2024) · Sensors · doi:10.3390/s24113579 · Cited by 21
“Wang et al. [37] addressed the problem of detecting small targets and dense occlusion on the surface of rails by introducing the SPD-Conv building block in YOLOv8 to improve detection attention to small and medium-sized targets, and the Focal-SIoU loss function was used to adjust the sample weights…”
MethodDGYOLOv8: An Enhanced Model for Steel Surface Defect Detection Based on YOLOv8
Zhu, G., Qi, H., & Lv, K. (2025) · Mathematics · doi:10.3390/math13050831 · Cited by 18
YOLO-VSI: An Improved YOLOv8 Model for Detecting Railway Turnouts Defects in Complex Environments
Yu, C., & Lu, Z. (2024) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2024.056413 · Cited by 19
MSConv-YOLO: An Improved Small Target Detection Algorithm Based on YOLOv8
Yang, L., & Asli, B. H. S. (2025) · Journal of Imaging · doi:10.3390/jimaging11080285 · Cited by 12
“…investigations and improvements, which can be categorised into the following three primary directions: Backbone Network Improvement: Wang et al. [31] enhanced YOLOv8’s backbone by integrating the SPD-Conv module into YOLOv8n, improving feature extraction for low-resolution images and small…”
Zhang, J., Zhang, R., Luan, F., & Zhang, H. (2025) · Applied Sciences · doi:10.3390/app15073588 · Cited by 11
Liu, Z., Peng, H., Ye, Z., Lian, C., Shen, H., Xiang, H., Xiong, B., & Mei, L. (2025) · Microscopy Research and Technique · doi:10.1002/jemt.24775 · Cited by 10
Radosavljevic, S., Rivero, A., Elouardi, A., & Rodríguez, S. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3645927 · Cited by 9
“In the past decade, researchers have rapidly adopted ML and DL for image analysis, driven by the increasing performance of CNNs [72].”
Chen, X., & Yu, C. (2025) · Multimedia Systems · doi:10.1007/s00530-024-01593-5 · Cited by 8
“Wang et al. [21] introduced a railway track defect detection algorithm based on the YOLOv8 model to address the challenges of detecting small and densely occluded objects in railway track surface defect detection.”
Real-time conveyor belt damage detection method based on improved YOLO
Hu, Y., Zhang, Y., Shen, J., & Wang, G. (2025) · Engineering Research Express · doi:10.1088/2631-8695/adc8fd · Cited by 8
A Multi-Modal Approach to Rail Surface Condition Analysis: The MFDF-Net
Shen, Y., Zhong, Q., Zheng, S., Li, L., & Peng, L. (2024) · IEEE Access · doi:10.1109/access.2024.3427347 · Cited by 8
“Wang et al. [40] proposed an improved algorithm based on the YOLOv8 model to enhance detection of small and densely occluded targets in railway surface defect detection.”
MethodStudy on Model Response of ACFM for Rail RCF Crack Based on Destructive Field Test
Wang, C., & Zhou, Y. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3520832 · Cited by 7
“NDT approaches, such as surface image analysis [18], magnetic particle inspection (MPI) [19], eddy current pulsed thermography (ECPT) [20], eddy current testing (ECT) [21], and ACFM [9], are investigated for rail RCF crack detection and evaluation.”
YOLOv8-SC: improving the YOLOv8 network for real-time detection of automotive coated surface defects
Ling, L., Zhu, C., Liu, M., Hu, J., Zhang, X., & Ge, M. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/adb05b · Cited by 8
Zhang, L., Chen, Y., Zheng, L., Yan, B., Zhang, J., Xie, A., & Lou, S. (2024) · Coatings · doi:10.3390/coatings14111466 · Cited by 6
Enhanced Ground Fissure Detection in Mining Areas Based on Visible–Infrared Image Fusion and YOLOv5
Zhao, Y., Zhao, L., Guo, J., Zhang, K., Ling, C., Wang, S., & Bian, H. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3552923 · Cited by 6
“This suggests that EMA may help optimize the performance of the AFPN neck [66], [67].”
Self-citationTrack fastener detection in special scenarios based on TSR-Net
Bai, T., Duan, J., Fu, H., & Zong, H. (2024) · Engineering Research Express · doi:10.1088/2631-8695/ad56fe · Cited by 6
LiMS-Net: Lightweight metal surface defect detection network
Zhu, Y., & Lin, Y. (2025) · Displays · doi:10.1016/j.displa.2025.103227 · Cited by 7
An improved lightweight YOLOv11 algorithm for weld surface defect detection
Zhang, R., Pan, C., Chen, Z., Fan, J., Chen, Z., & Yuan, B. (2026) · Scientific Reports · doi:10.1038/s41598-026-41568-2 · Cited by 5
Target Detection of Diamond Nanostructures Based on Improved YOLOv8 Modeling
Guo, F., Guo, X., Guo, L., Wang, Y., Wang, Q., Liu, S., Zhang, M., Zhang, L., & Gai, Z. (2024) · Nanomaterials · doi:10.3390/nano14131115 · Cited by 5
“Compared to the outstanding models of the previous YOLO series, such as YOLOv5 and YOLOv7, YOLOv8 is characterized by higher detection accuracy and speed [24].”
ResultSelf-citationMaizeStar-YOLO: Precise Detection and Localization of Seedling-Stage Maize
Chu, T., Zha, H., Wang, Y., Yao, Z., Wang, X., Wu, C., & Liao, J. (2025) · Agronomy · doi:10.3390/agronomy15081788 · Cited by 5
Self-citationVadukkal, U. K. V., Cardellicchio, A., Nitti, M., Fiume, M., & Renò, V. (2025) · IEEE Access · doi:10.1109/access.2025.3634804 · Cited by 4
“Other authors focus on the rail surface, using RGB cameras for detecting defects such as cracks, scars, breaks [55], [56], or surface failure patterns [34].”
Ji, W., Fang, H., Xu, B., Wu, K., & Xie, L. (2025) · Solar Energy · doi:10.1016/j.solener.2025.114166 · Cited by 5
Hu, Z., Lau, A., Tao, R., (6908), Y. L., & Pan, Y. (2026) · Journal of Railway Science and Technology · doi:10.1016/j.jrst.2026.05.005 · Cited by 4
YOLOv8-UCB: Visual Detection of Pouch Battery Using Improved YOLOv8
Hao, H., & Yu, X. (2024) · IEEE Access · doi:10.1109/access.2024.3520181 · Cited by 4
Lu, W., Zheng, H., Xiao, S., Xue, W., & Yang, S. (2025) · Journal of Nondestructive Evaluation · doi:10.1007/s10921-025-01235-4 · Cited by 5
“Similarly, Wang et al. [30] integrated an efficient multi-scale attention (EMA) into the neck component to enhance the model’s ability to utilize features from differ - ent layers, achieving a 5% improvement in mAP@50% over the base model.”
Multi-Scale Plastic Lunch Box Surface Defect Detection Based on Dynamic Convolution
Yang, J., Zhang, G., Ge, Y., Jingzhuo, S., Wang, Y., & Li, J. (2024) · IEEE Access · doi:10.1109/access.2024.3450719 · Cited by 3
“YOLOv8 introduces new feature flow module, new detection header and frame matching strategy, which significantly improves model speed and reduces model size [20].”
BackgroundSelf-citationMulti-Scale Feature Similarity and Object Detection for Small Printing Defects Detection
Lou, H., Zheng, Y., Chen, W., & Liu, H. (2024) · IEEE Access · doi:10.1109/access.2024.3521403 · Cited by 3
“The mainstream object detection algorithms (Efficientdet [34], SSD [35], YOLOv3 [11], YOLOv4 [36], YOLOv5 [37], YOLOv8 [38], [39], YOLOv11 and Faster R-CNN [40]) are compared with the detection accuracy of the proposed model on the printing defect dataset.”
Zhang, Z., Zhang, L., Lu, X., Ma, T., Huang, F., & Zhong, S. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2025.1711309 · Cited by 3
“…(R-FCN)(Tang et al., 2020), and spatial pyramid pooling networks (SPPNet) (Han et al., 2020), and end-to-end detection frameworks such as single shot multibox detector (SSD) (Bai et al., 2021), You Only Look Once (YOLO) (Wang et al., 2024), and detection transformer (DETR) (Gibert et al., 2016).”
Lester, D., Zhu, P., Narman, H. S., & Alzarrad, A. (2025) · ASEE North Central Section Conference · doi:10.18260/1-2--54646 · Cited by 5
EN-YOLO: a featured-enhanced model for precise hot rolled steel surface defect detection
Hao, Z., Liu, B., & Xu, B. (2025) · Nondestructive Testing And Evaluation · doi:10.1080/10589759.2025.2524604 · Cited by 4
“The EN-YOLO is compared with several state-of-the-art models, including SSD, YOLOv5l [56], YOLOv5x, YOLOv7, YOLOv8m [57], YOLOv9s [58], YOLOv10b [59], YOLOv11m [60], YOLOv12m [61] and RT-DETR [53].”
Wang, J., Zheng, X., Liu, K., Niu, F., Zhang, D., Lin, Y., & Chen, J. (2025) · Engineering Research Express · doi:10.1088/2631-8695/add1ab · Cited by 3
Rappe, N. A., Kirda, A. W., Yassin, H., Bartoszuk, M., & Caesarendra, W. (2025) · Advances in Science and Technology – Research Journal · doi:10.12913/22998624/210216 · Cited by 2
Damage assessment of bridges using segmentation and shape analysis from aerial imagery
Kubo, S., Zheng, S., & Chun, P. (2025) · Proceedings of the Institution of Civil Engineers - Forensic Engineering · doi:10.1680/jfoen.25.00025 · Cited by 2
Chandramouli, A., Song, H., Liu, M., Damai, A., Narman, H. S., & Alzarrad, A. (2025) · IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · doi:10.1109/uemcon67449.2025.11267623 · Cited by 2
“Another commonality between both studies is modifying YOLO’s default loss function, substituting Alpha-IoU and Focal-SIoU, respectively, to improve robustness against hard-to-classify samples [13], [14].”
InfluentialYu, L., Feng, X., Zeng, Y., Guo, W., Yang, X., Zhang, X., Tan, Y., Sun, C., Lu, X., & Sun, H. (2025) · Electronics · doi:10.3390/electronics15010097 · Cited by 3
Research on a Wear Defect Detection Method for a Switch Sliding Baseplate Based on Improved Yolov5
Jiang, Q., Gao, R., Zhao, Y., Yu, W., Dang, Z., & Deng, S. (2024) · Lubricants · doi:10.3390/lubricants12120422 · Cited by 2
RDX-Net: an efficient model for rail defect detection
Jiang, Z. (2025) · Engineering Research Express · doi:10.1088/2631-8695/ade94e · Cited by 3
Chung, M., Lin, C.-W., Lin, J.-W., Lin, C.-C., Gao, C.-Y., Chen, P.-C., & Ma, Y.-X. (2025) · Sensors and Materials · doi:10.18494/sam5487 · Cited by 1
Duan, Q., & Han, P. (2025) · International Conference on Computer Information Science and Application Technology (CISAT) · doi:10.1109/cisat66811.2025.11181953 · Cited by 1
“Therefore, developing advanced object detection algorithms that can efficiently address these challenges has become a research hotspot [3].”
Improving the Accuracy of Subway Mobile Laser Scanning Point Clouds Based on Target Constraints
Liu, H., Wu, H., Yao, L., Yi, K., Xu, Z., Ding, X., & Yang, K. (2025) · Journal of Surveying Engineering · doi:10.1061/jsued2.sueng-1625 · Cited by 1
A TTDCNN-based condition monitoring and crack depth severity assessment of railway wheel axle defect
Dhanapaul, A. L., & Nandakumar, P. (2025) · Nondestructive Testing And Evaluation · doi:10.1080/10589759.2025.2559063 · Cited by 1
“Wang et al. [24] utilised an enhanced YOLOv8 technique to detect the rainway surface defects.”
Railway rail defect detection based on improved YOLOv11 algorithm
Chung, M., & Zhang, C.-Y. (2025) · IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan) · doi:10.1109/icce-taiwan66881.2025.11207840 · Cited by 1
YOLOv8-BPSNB: an improved YOLOv8-based algorithm for blueberry fruit ripeness detection
Zhu, Y., Chen, J., San, H., Zhang, F., & Zu, Y. (2026) · Journal of Food Measurement & Characterization · doi:10.1007/s11694-026-04616-z · Cited by 1
Bai, T., Wang, X., Wang, Y., Li, W., & Xiang, H. (2025) · The Journal of Supercomputing · doi:10.1007/s11227-025-08155-0 · Cited by 1
Self-citationA lightweight multi-wedge belt defect detection method combining CNN and transformer
Wang, J., Zheng, X., Liu, Z., Niu, F., Liu, K., Zhang, D., Lin, Y., & Lu, C. (2026) · Engineering Research Express · doi:10.1088/2631-8695/ae3458 · Cited by 1
Enhanced Traffic Safety by Road Anomaly Detection Using YOLOv8 and Machine Learning Models
Dhake, D., Prakash, G., Ingle, R., Nain, A., Kadam, K., Gite, S., Pareek, P., & Pradhan, B. (2026) · Smart sensors, measurement and instrumentation · doi:10.1007/978-3-032-05507-1_19 · Cited by 1
Sansyzbay, K., Bakhtiyarova, Y., Turgambay, Y., Тасболатова, Л., Kismanova, A., & Zhumagul, A. (2026) · Automation · doi:10.3390/automation7030071 · Cited by 1
He, L., Shen, H., Wei, Q., Su, Z., Shi, H., Liu, Y., & Wang, K. (2026) · Measurement · doi:10.1016/j.measurement.2026.122477 · Cited by 1
TrackDef-YOLO: An Improved YOLOv11 Model for Railway Track Surface Defect Detection
李, 梦. (2025) · Artificial Intelligence and Robotics Research · doi:10.12677/airr.2025.141021 · Cited by 1
Fault Diagnosis of Tobacco Packaging Machine Detection using You Only Look Once Version 8
Ran, B., Bu, J., & Han, S. (2024) · International Conference on Integrated Intelligence and Communication Systems (ICIICS) · doi:10.1109/iciics63763.2024.10859929
Salsabila, P. M., Luthfiarta, A., Nugraha, A., Muttaqin, A. N. I., & Zarifa, Y. (2024) · Building of Informatics Technology and Science (BITS) · doi:10.47065/bits.v6i3.6225
“Dengan cara ini, YOLOv8 tidak hanya cepat, tetapi juga lebih akurat karena mempertimbangkan konteks gambar secara keseluruhan [16].”
Oracle recognition based on improved yolov8
Duan, X., & Li, Q. (2024) · International Conference on Image Processing, Multimedia Technology and Machine Learning (IPMML) · doi:10.1145/3722405.3722433
A detection method for surface defects on metal workpieces based on feature enhancement
Liu, J., Le, W., Jia, Z., & Tian, J. (2025) · International Conference on Software Engineering and Computer Applications · doi:10.1145/3747912.3747943
“Wang et al. [6] applied multi-scale attention in YOLOv8, which adjusts channel weights and captures cross-dimensional pixel relations.”
Enhancing rail track defect detection using an ensemble approach with self-attention mechanism
Bhat, S., & Karegowda, A. G. (2025) · Proceedings of the Institution of Mechanical Engineers Part F Journal of Rail and Rapid Transit · doi:10.1177/09544097251378313
Road pothole recognition in multiple weather scenarios based on YOLOv8 optimization algorithm
Liu, P., Ma, R., & Li, M. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/ae2caf
Enhanced YOLOv8n-Based Approach for Traffic Object Detection under Complex Environmental Conditions
Shi, J., & Liu, X. (2025) · International Conference on Image Processing, Computer Vision and Machine Learning (ICICML) · doi:10.1109/icicml67980.2025.11333668
“Traditional Feature Pyramid Networks (FPN [10]) aggregate multi-scale features in a top-down manner but are constrained by unidirectional information flow.”
MSI-YOLO: An Improved YOLOv8n Algorithm for PCB Surface Defect Detection
Li, Y., & Wen, T. (2025) · International Conference on Electronic Information Engineering and Computer Technology (EIECT) · doi:10.1109/eiect68017.2025.11331785
Implementation of an IoT-Based Smart Railway Track Inspection and Collision Mitigation
Suneel, E., Bhavitha, V., Harshini, V., Majeed, S. A., & Gopi, V. (2025) · IEEE International Conference on Information Technology, Electronics and Intelligent Communication Systems (ICITEICS) · doi:10.1109/iciteics64870.2025.11340863
“The goal of future research is to improve the network architecture and make it possible to deploy edge computing for practical uses [6].”
PMB-DETR: A Multi-Scale and Multi-Branch Enhanced Network for Rail Surface Defect Detection
Wang, L., Deng, L., Fan, X., & Tang, Y. (2025) · International Conference on Sensing, Diagnostics, Prognostics, and Control (SDPC) · doi:10.1109/sdpc68151.2025.11347947
“Wang Y. et al. [6], addressing the problem of small and densely occluded targets in track surface defect detection, proposed a YOLOv8-based railway track defect detection model.”
Self-citationLi, Z., Guo, B., Liu, X., Wu, S., Zhang, J., Wang, S., & Sun, H. (2026) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2026.3659686
“Similarly, although computer vision techniques [19] e ff ectively detect surface defects, they lack robustness against the intense variable illumination and reflective noise characteristic of open heavy-haul environments.”
Bridge Surface Defect Detection Based on Machine Learning
Zhang, H., & Chen, H. (2025) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio66223.2025.11376191
Design and Study of Three-Dimensional Pipeline Defect Detection Robot
Li, S., Xu, L., Kang, Z., Lv, Z., & Wang, Z. (2025) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio66223.2025.11378318
“The detected defect images are transferred to the computer side and trained by YOLOV8 target detection macromodel, which improves the accuracy and efficiency of defect detection[12].”
ASL-YOLO: A Lightweight and Efficient YOLOv11 Model for Rail Surface Defect Detection
Lin, Y., & Mou, L. (2025) · International Conference on Signal Processing, Computer Networks and Communications (SPCNC) · doi:10.1109/spcnc68200.2025.11406249
“Focusing on small-defect perception, reference [7] improves YOLOv8n by adding SPD-Conv blocks in the backbone, an EMA attention module in the neck, and a Focal-SIoU loss to enhance detection of small and occluded defects without increasing model size.”
Road traffic vehicle recognition and detection based on the optimized YOLOv8 algorithm
Chen, H. (2025) · International Conference on Signal Processing, Computer Networks and Communications (SPCNC) · doi:10.1109/spcnc68200.2025.11406426
“…aggregates features at multiple scales through three parallel 5×5 Max pooling layers and one join operation, expands the receptive field, enhances the global information capture ability, and improves the computational efficiency by approximately 30% compared with the traditional SPP [6].”
AGBR-YOLO: an infrared crack detection algorithm based on improved YOLOv11
(328526), H. Y., Zhao, D., He, M., Hou, G., Lele, S., Zhang, R., & He, J. (2026) · Engineering Research Express · doi:10.1088/2631-8695/ae4da7
Multi-class rail defect detection using advanced artificial intelligence
Mordia, R., & Verma, A. K. (2026) · High-speed Railway · doi:10.1016/j.hspr.2026.03.002
RailNet: Railway Track Anomaly Detection via Image Processing With Hybrid Deep Learning Techniques
Saeed, U., Waqas, M., Shah, A. A., Laghari, A. A., Almadhor, A. S., Kryvinska, N., Rajab, R., Laghari, R. A., & Kausar, S. (2026) · IET Image Processing · doi:10.1049/ipr2.70361
Wang, W., Fan, W., Dong, F., Zeng, B., Yu, W., & Deng, X. (2026) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2026.078416
Enhancing Rail Surface Defect Detection Using a Hybrid YOLOv11–Vision Transformer Framework
Markapudi, B., Gundiga, L., Kondeti, V. R. B., Kanumuri, S., & Maddala, Y. (2026) · Journal of Trends in Computer Science and Smart Technology · doi:10.36548/jtcsst.2026.2.005
“Enhanced YOLOv8 algorithms provide a more advanced feature extraction process and fusion method, which increases the detection efficiency of rail surface defects in complicated railway scenarios [7].”
YOLO11-based Defect Detection for Wafer Maps in MIR-WM811K Dataset
Li, J. W., Wang, X., & Chen, J. P. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11487401
“The YOLO series, known for its superior speed-accuracy balance, has become a popular choice [8].”
Infrared-Visible Image Fusion for Transformer Bushing Inspection Using Enhanced YOLOv8
Ye, Z., Yang, S., Guo, B., Zhang, X., Xu, Z., Huang, J., & Zhao, S. (2025) · Asian Conference on Artificial Intelligence Technology (ACAIT) · doi:10.1109/acait67930.2025.11522639
Multi-modal fusion rail surface defect detection based on expert decision-making
Zhang, C., Tian, A., Zhang, G., Cheng, Y., & Wang, H. (2026) · Measurement Science and Technology · doi:10.1088/1361-6501/ae6c44
Li, J. (2026) · Academic Journal of Emerging Technologies · doi:10.63313/ajet.9058
An Improved Method for Chip Defect Detection Based on YOLOv8
Kang, B., Chen, J. C., & An, X. (2026) · IEEE International Conference on Control, Electronics and Computer Technology (ICCECT) · doi:10.1109/iccect68671.2026.11565375
“…through the optimization of loss functions and the application of attention mechanisms, has improved detection accuracy to some extent, but there is still a lack of systematic exploration of the collaborative design between gradient optimization strategies and multi-scale feature fusion [9].”
Online surface defect detection of steel rails using a multi-module optimized YOLO framework
Cao, X., Yang, X., Wang, S., Xu, K., Zhu, H., & Jiao, J. (2026) · Applied Soft Computing · doi:10.1016/j.asoc.2026.116155
AWT ‐YOLO: Lightweight Rail Surface Defect Detection Algorithm Based on YOLOv11 Improvement
Liu, L., Sun, Y., Wang, J., Wang, Z., & Hu, C. (2026) · Engineering Reports · doi:10.1002/eng2.71001
“For instance, Wang [ 14] showed that a lightweight backbone modification can maintain high detection performance while achieving a detection accuracy of 93.9% and an average precision of 94.1%.”
InfluentialComparative Analysis of ResNet50 and MobileNetV2 for Multiclass Railway Track Defect Classification
Toshniwal, P., Tandon, P., Gadavi, S., Jisy, N. K., & Srikanya, D. (2026) · International Conference on Advanced Computing and Knowledge Engineering (ICACKE) · doi:10.1109/icacke65470.2026.11651230
YOLOv11-ODConv for Cross-Domain Micro-Scratch Detection in Photovoltaic Glass Manufacturing
Bence, N., & Mihály, P. (2026) · Journal of Applied Automation Technologies · doi:10.64972/jaat.2026v4.388p33e:442-457
“At a high throughput, manual examination is unable to maintain consistent sensitivity [6].”
Kumar, A., Kushwaha, D. S., & Harsha, S. P. (2026) · Journal of Vibration Engineering & Technologies · doi:10.1007/s42417-026-02727-x
Dhanapaul, A. L., & Nandakumar, P. (2026) · Fatigue & Fracture of Engineering Materials & Structures · doi:10.1111/ffe.70395
Performance Evaluation Of Yolov8 For Vision Based Multi Class Rail Surface Damage Detection
Arifianto, T., Sunaryo, S., Arifidin, M. A. A., Muqarrabiiansyah, M. R., Prasetyo, G., & Puspitarini, E. W. (2026) · Jurnal Penelitian Sekolah Tinggi Transportasi Darat · doi:10.55511/jpsttd.v17i1.857
Xu, G., Xie, J., & Chao, Q. (2026) · Research on Engineering Structures and Materials · doi:10.17515/resm2026-1850ai0710rs
Dynamic Focal Loss adjustment for railway defect detection
Koç, M., Özdemir, R., & Gerek, Ö. N. (2026) · TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES · doi:10.55730/1300-0632.4276
Korkmaz, D., & Açıkgöz, H. (2026) · Applied Soft Computing · doi:10.1016/j.asoc.2026.116573
DB-YOLO: An Improved YOLOV8 Model Based on The Second Backbone for Small Defects on PCB Surface
Liu, J., Gong, Y., Zheng, H., Cai, H., & Yang, J. (2025) · Journal of Research in Multidisciplinary Methods and Applications
Abisheva, G., Shamshanovna, R. B., Aidynov, T., & Saparbek, M. (2026) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2026.088467
“…mechanism to boost the detection results of small defects on the rail surface [11]; Luo et al. integrated an improved YOLOv5s with a Swin-Transformer-augmented backbone [12]; and Wang et al. proposed an improved YOLOv8 detector to enhance the detection results of defects on the rail surface [13].”
Budak, İ., Bal, S., & Korkmaz, H. (2026) · Applied Sciences · doi:10.3390/app162110965
Metadata
- Authors
- Wang, Y., Zhang, K., Wang, L., & Wu, L.
- Affiliations
- Zhejiang Normal University (CN)
- Venue
- IEEE Access
- Published
- 2024
- Volume
- 12
- Pages
- 44984-44997
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W4393170784
- Cited by
- 82
- References
- 33
- Citations by year
- 2026: 23 · 2025: 39 · 2024: 20
- FWCI
- 13.69
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- IEEE Access (publishedVersion, cc-by-nc-nd)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Science and Technology Planning Project of Jinhua (2022-1-096)
- Topics
- Physical Sciences › Engineering › Civil and Structural Engineering › Infrastructure Maintenance and Monitoring
- Physical Sciences › Engineering › Industrial and Manufacturing Engineering › Industrial Vision Systems and Defect Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- rail surface defect detection, YOLOv8, small object detection, occluded object detection, SPD-Conv, railway track inspection, object detection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.