DP-YOLO: A Lightweight Real-Time Detection Algorithm for Rail Fastener Defects
To enable accurate and efficient real-time detection of rail fastener defects under resource-constrained environments, we propose DP-YOLO, an advanced lightweight algorithm based on YOLOv5s with four key optimizations. First, we design a Depthwise Separable Convolution Stage Partial (DSP) module tha...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/7/2139 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849769513666478080 |
|---|---|
| author | Lihua Chen Qi Sun Ziyang Han Fengwen Zhai |
| author_facet | Lihua Chen Qi Sun Ziyang Han Fengwen Zhai |
| author_sort | Lihua Chen |
| collection | DOAJ |
| description | To enable accurate and efficient real-time detection of rail fastener defects under resource-constrained environments, we propose DP-YOLO, an advanced lightweight algorithm based on YOLOv5s with four key optimizations. First, we design a Depthwise Separable Convolution Stage Partial (DSP) module that integrates depthwise separable convolution with a CSP residual connection strategy, reducing model parameters while enhancing recognition accuracy. Second, we introduce a Position-Sensitive Channel Attention (PSCA) mechanism, which calculates spatial statistics (mean and standard deviation) across height and width dimensions for each channel feature map. These statistics are multiplied across corresponding dimensions to generate channel-specific weights, enabling dynamic feature recalibration. Third, the Neck network adopts a GhostC3 structure, which reduces redundancy through linear operations, further minimizing computational costs. Fourth, to improve multi-scale adaptability, we replace the standard loss function with Alpha-IoU, enhancing model robustness. Experiments on the augmented Roboflow Universe Fastener-defect-detection Dataset demonstrate DP-YOLO’s effectiveness: it achieves 87.1% detection accuracy, surpassing the original YOLOv5s by 1.3% in mAP0.5 and 2.1% in mAP0.5:0.95. Additionally, the optimized architecture reduces parameters by 1.3% and computational load by 15.19%. These results validate DP-YOLO’s practical value for resource-efficient, high-precision defect detection in railway maintenance systems. |
| format | Article |
| id | doaj-art-696d2307afe84ceda4d49fbcaacbbfb0 |
| institution | DOAJ |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-696d2307afe84ceda4d49fbcaacbbfb02025-08-20T03:03:24ZengMDPI AGSensors1424-82202025-03-01257213910.3390/s25072139DP-YOLO: A Lightweight Real-Time Detection Algorithm for Rail Fastener DefectsLihua Chen0Qi Sun1Ziyang Han2Fengwen Zhai3School of Information Science & Technology, Southwest Jiaotong University, Chengdu 611756, ChinaSchool of Electronics and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, ChinaSchool of Electronics and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, ChinaSchool of Electronics and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, ChinaTo enable accurate and efficient real-time detection of rail fastener defects under resource-constrained environments, we propose DP-YOLO, an advanced lightweight algorithm based on YOLOv5s with four key optimizations. First, we design a Depthwise Separable Convolution Stage Partial (DSP) module that integrates depthwise separable convolution with a CSP residual connection strategy, reducing model parameters while enhancing recognition accuracy. Second, we introduce a Position-Sensitive Channel Attention (PSCA) mechanism, which calculates spatial statistics (mean and standard deviation) across height and width dimensions for each channel feature map. These statistics are multiplied across corresponding dimensions to generate channel-specific weights, enabling dynamic feature recalibration. Third, the Neck network adopts a GhostC3 structure, which reduces redundancy through linear operations, further minimizing computational costs. Fourth, to improve multi-scale adaptability, we replace the standard loss function with Alpha-IoU, enhancing model robustness. Experiments on the augmented Roboflow Universe Fastener-defect-detection Dataset demonstrate DP-YOLO’s effectiveness: it achieves 87.1% detection accuracy, surpassing the original YOLOv5s by 1.3% in mAP0.5 and 2.1% in mAP0.5:0.95. Additionally, the optimized architecture reduces parameters by 1.3% and computational load by 15.19%. These results validate DP-YOLO’s practical value for resource-efficient, high-precision defect detection in railway maintenance systems.https://www.mdpi.com/1424-8220/25/7/2139rail fastener defects detectionlightweightYOLOv5sattention mechanismstatistical information weighted feature maps |
| spellingShingle | Lihua Chen Qi Sun Ziyang Han Fengwen Zhai DP-YOLO: A Lightweight Real-Time Detection Algorithm for Rail Fastener Defects Sensors rail fastener defects detection lightweight YOLOv5s attention mechanism statistical information weighted feature maps |
| title | DP-YOLO: A Lightweight Real-Time Detection Algorithm for Rail Fastener Defects |
| title_full | DP-YOLO: A Lightweight Real-Time Detection Algorithm for Rail Fastener Defects |
| title_fullStr | DP-YOLO: A Lightweight Real-Time Detection Algorithm for Rail Fastener Defects |
| title_full_unstemmed | DP-YOLO: A Lightweight Real-Time Detection Algorithm for Rail Fastener Defects |
| title_short | DP-YOLO: A Lightweight Real-Time Detection Algorithm for Rail Fastener Defects |
| title_sort | dp yolo a lightweight real time detection algorithm for rail fastener defects |
| topic | rail fastener defects detection lightweight YOLOv5s attention mechanism statistical information weighted feature maps |
| url | https://www.mdpi.com/1424-8220/25/7/2139 |
| work_keys_str_mv | AT lihuachen dpyoloalightweightrealtimedetectionalgorithmforrailfastenerdefects AT qisun dpyoloalightweightrealtimedetectionalgorithmforrailfastenerdefects AT ziyanghan dpyoloalightweightrealtimedetectionalgorithmforrailfastenerdefects AT fengwenzhai dpyoloalightweightrealtimedetectionalgorithmforrailfastenerdefects |