Railway Tracks Extraction from High Resolution Unmanned Aerial Vehicle Images Using Improved NL-LinkNet Network
The accurate detection of railway tracks from unmanned aerial vehicle (UAV) images is essential for intelligent railway inspection and the development of electronic railway maps. Traditional computer vision algorithms struggle with the complexities of high-precision track extraction due to challenge...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-10-01
|
| Series: | Drones |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-446X/8/11/611 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850217082859290624 |
|---|---|
| author | Jing Wang Xiwei Fan Yunlong Zhang Xuefei Zhang Zhijie Zhang Wenyu Nie Yuanmeng Qi Nan Zhang |
| author_facet | Jing Wang Xiwei Fan Yunlong Zhang Xuefei Zhang Zhijie Zhang Wenyu Nie Yuanmeng Qi Nan Zhang |
| author_sort | Jing Wang |
| collection | DOAJ |
| description | The accurate detection of railway tracks from unmanned aerial vehicle (UAV) images is essential for intelligent railway inspection and the development of electronic railway maps. Traditional computer vision algorithms struggle with the complexities of high-precision track extraction due to challenges such as diverse track shapes, varying angles, and complex background information in UAV images. While deep learning neural networks have shown promise in this domain, they still face limitations in precisely extracting track line edges. To address these challenges, this paper introduces an improved NL-LinkNet network, named NL-LinkNet-SSR, designed specifically for railway track detection. The proposed NL-LinkNet-SSR integrates a Sobel edge detection module and a SimAM attention module to enhance the model’s accuracy and robustness. The Sobel edge detection module effectively captures the edge information of track lines, improving the segmentation and extraction of target edges. Meanwhile, the parameter-free SimAM attention module adaptively emphasizes significant features while suppressing irrelevant information, broadening the model’s perceptual field and improving its responsiveness to target areas. Experimental results show that the NL-LinkNet-SSR significantly outperforms the original NL-LinkNet model across multiple key metrics, including a more than 0.022 increase in accuracy, over a 4% improvement in F1-score, and a more than 3.5% rise in mean Intersection over Union (mIoU). These enhancements suggest that the improved NL-LinkNet-SSR offers a more reliable solution for railway track detection, advancing the field of intelligent railway inspection. |
| format | Article |
| id | doaj-art-c2f39c78a1b64dfb9c5304a4e2eebc3f |
| institution | OA Journals |
| issn | 2504-446X |
| language | English |
| publishDate | 2024-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Drones |
| spelling | doaj-art-c2f39c78a1b64dfb9c5304a4e2eebc3f2025-08-20T02:08:09ZengMDPI AGDrones2504-446X2024-10-0181161110.3390/drones8110611Railway Tracks Extraction from High Resolution Unmanned Aerial Vehicle Images Using Improved NL-LinkNet NetworkJing Wang0Xiwei Fan1Yunlong Zhang2Xuefei Zhang3Zhijie Zhang4Wenyu Nie5Yuanmeng Qi6Nan Zhang7Key Laboratory of Seismic and Volcanic Hazards, China Earthquake Administration, Beijing 100029, ChinaKey Laboratory of Seismic and Volcanic Hazards, China Earthquake Administration, Beijing 100029, ChinaChina Railway Design Corporation, Tianjin 300308, ChinaLand Satellite Remote Sensing Application Center, Ministry of Natural Resources, Beijing 100034, ChinaInstitute of Strategic Planning, Chinese Academy of Environmental Planning, Beijing 100012, ChinaKey Laboratory of Seismic and Volcanic Hazards, China Earthquake Administration, Beijing 100029, ChinaKey Laboratory of Seismic and Volcanic Hazards, China Earthquake Administration, Beijing 100029, ChinaKey Laboratory of Seismic and Volcanic Hazards, China Earthquake Administration, Beijing 100029, ChinaThe accurate detection of railway tracks from unmanned aerial vehicle (UAV) images is essential for intelligent railway inspection and the development of electronic railway maps. Traditional computer vision algorithms struggle with the complexities of high-precision track extraction due to challenges such as diverse track shapes, varying angles, and complex background information in UAV images. While deep learning neural networks have shown promise in this domain, they still face limitations in precisely extracting track line edges. To address these challenges, this paper introduces an improved NL-LinkNet network, named NL-LinkNet-SSR, designed specifically for railway track detection. The proposed NL-LinkNet-SSR integrates a Sobel edge detection module and a SimAM attention module to enhance the model’s accuracy and robustness. The Sobel edge detection module effectively captures the edge information of track lines, improving the segmentation and extraction of target edges. Meanwhile, the parameter-free SimAM attention module adaptively emphasizes significant features while suppressing irrelevant information, broadening the model’s perceptual field and improving its responsiveness to target areas. Experimental results show that the NL-LinkNet-SSR significantly outperforms the original NL-LinkNet model across multiple key metrics, including a more than 0.022 increase in accuracy, over a 4% improvement in F1-score, and a more than 3.5% rise in mean Intersection over Union (mIoU). These enhancements suggest that the improved NL-LinkNet-SSR offers a more reliable solution for railway track detection, advancing the field of intelligent railway inspection.https://www.mdpi.com/2504-446X/8/11/611deep learningedge detectionrailway track detectionattention mechanism |
| spellingShingle | Jing Wang Xiwei Fan Yunlong Zhang Xuefei Zhang Zhijie Zhang Wenyu Nie Yuanmeng Qi Nan Zhang Railway Tracks Extraction from High Resolution Unmanned Aerial Vehicle Images Using Improved NL-LinkNet Network Drones deep learning edge detection railway track detection attention mechanism |
| title | Railway Tracks Extraction from High Resolution Unmanned Aerial Vehicle Images Using Improved NL-LinkNet Network |
| title_full | Railway Tracks Extraction from High Resolution Unmanned Aerial Vehicle Images Using Improved NL-LinkNet Network |
| title_fullStr | Railway Tracks Extraction from High Resolution Unmanned Aerial Vehicle Images Using Improved NL-LinkNet Network |
| title_full_unstemmed | Railway Tracks Extraction from High Resolution Unmanned Aerial Vehicle Images Using Improved NL-LinkNet Network |
| title_short | Railway Tracks Extraction from High Resolution Unmanned Aerial Vehicle Images Using Improved NL-LinkNet Network |
| title_sort | railway tracks extraction from high resolution unmanned aerial vehicle images using improved nl linknet network |
| topic | deep learning edge detection railway track detection attention mechanism |
| url | https://www.mdpi.com/2504-446X/8/11/611 |
| work_keys_str_mv | AT jingwang railwaytracksextractionfromhighresolutionunmannedaerialvehicleimagesusingimprovednllinknetnetwork AT xiweifan railwaytracksextractionfromhighresolutionunmannedaerialvehicleimagesusingimprovednllinknetnetwork AT yunlongzhang railwaytracksextractionfromhighresolutionunmannedaerialvehicleimagesusingimprovednllinknetnetwork AT xuefeizhang railwaytracksextractionfromhighresolutionunmannedaerialvehicleimagesusingimprovednllinknetnetwork AT zhijiezhang railwaytracksextractionfromhighresolutionunmannedaerialvehicleimagesusingimprovednllinknetnetwork AT wenyunie railwaytracksextractionfromhighresolutionunmannedaerialvehicleimagesusingimprovednllinknetnetwork AT yuanmengqi railwaytracksextractionfromhighresolutionunmannedaerialvehicleimagesusingimprovednllinknetnetwork AT nanzhang railwaytracksextractionfromhighresolutionunmannedaerialvehicleimagesusingimprovednllinknetnetwork |