Improving the Accuracy of Lane Detection by Enhancing the Long-Range Dependence
Lane detection is a common task in computer vision that involves identifying the boundaries of lanes on a road from an image or a video. Improving the accuracy of lane detection is of great help to advanced driver assistance systems and autonomous driving that help cars to identify and keep in the c...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/11/2518 |
| _version_ | 1797597686837280768 |
|---|---|
| author | Bo Liu Li Feng Qinglin Zhao Guanghui Li Yufeng Chen |
| author_facet | Bo Liu Li Feng Qinglin Zhao Guanghui Li Yufeng Chen |
| author_sort | Bo Liu |
| collection | DOAJ |
| description | Lane detection is a common task in computer vision that involves identifying the boundaries of lanes on a road from an image or a video. Improving the accuracy of lane detection is of great help to advanced driver assistance systems and autonomous driving that help cars to identify and keep in the correct lane. Current high-accuracy models of lane detection are mainly based on artificial neural networks. Among them, CLRNet is the latest famous model, which attains high lane detection accuracy. However, in some scenarios, CLRNet attains lower lane detection accuracy, and we revealed that this is caused by insufficient global dependence information. In this study, we enhanced CLRNet and proposed a new model called NonLocal CLRNet (NLNet). NonLocal is an algorithmic mechanism that captures long-range dependence. NLNet employs NonLocal to acquire more long-range dependence information or global information and then applies the acquired information to a Feature Pyramid Network (FPN) in CLRNet for improving lane detection accuracy. Using the CULane dataset, we trained NLNet. The experimental results showed that NLNet outperformed state-of-the-art models in terms of accuracy in most scenarios, particularly in the no-line scenario and night scenario. This study is very helpful for developing more accurate lane detection models. |
| first_indexed | 2024-03-11T03:09:05Z |
| format | Article |
| id | doaj.art-ea74ff1b996c4618b4dfc8385d2b8cec |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T03:09:05Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-ea74ff1b996c4618b4dfc8385d2b8cec2023-11-18T07:46:02ZengMDPI AGElectronics2079-92922023-06-011211251810.3390/electronics12112518Improving the Accuracy of Lane Detection by Enhancing the Long-Range DependenceBo Liu0Li Feng1Qinglin Zhao2Guanghui Li3Yufeng Chen4School of Computer Science and Engineering, Macau University of Science and Technology, Macau 999078, ChinaSchool of Computer Science and Engineering, Macau University of Science and Technology, Macau 999078, ChinaSchool of Computer Science and Engineering, Macau University of Science and Technology, Macau 999078, ChinaSchool of Artificial Intelligence and Computer, Science, Jiangnan University, Wuxi 214122, ChinaInstitute of Vehicle Information Control and Network Technology, Hubei University of Automotive Technology, Shiyan 442002, ChinaLane detection is a common task in computer vision that involves identifying the boundaries of lanes on a road from an image or a video. Improving the accuracy of lane detection is of great help to advanced driver assistance systems and autonomous driving that help cars to identify and keep in the correct lane. Current high-accuracy models of lane detection are mainly based on artificial neural networks. Among them, CLRNet is the latest famous model, which attains high lane detection accuracy. However, in some scenarios, CLRNet attains lower lane detection accuracy, and we revealed that this is caused by insufficient global dependence information. In this study, we enhanced CLRNet and proposed a new model called NonLocal CLRNet (NLNet). NonLocal is an algorithmic mechanism that captures long-range dependence. NLNet employs NonLocal to acquire more long-range dependence information or global information and then applies the acquired information to a Feature Pyramid Network (FPN) in CLRNet for improving lane detection accuracy. Using the CULane dataset, we trained NLNet. The experimental results showed that NLNet outperformed state-of-the-art models in terms of accuracy in most scenarios, particularly in the no-line scenario and night scenario. This study is very helpful for developing more accurate lane detection models.https://www.mdpi.com/2079-9292/12/11/2518lane detectionlong-range dependenceCLRNetnonlocal mechanismfeature pyramid network |
| spellingShingle | Bo Liu Li Feng Qinglin Zhao Guanghui Li Yufeng Chen Improving the Accuracy of Lane Detection by Enhancing the Long-Range Dependence Electronics lane detection long-range dependence CLRNet nonlocal mechanism feature pyramid network |
| title | Improving the Accuracy of Lane Detection by Enhancing the Long-Range Dependence |
| title_full | Improving the Accuracy of Lane Detection by Enhancing the Long-Range Dependence |
| title_fullStr | Improving the Accuracy of Lane Detection by Enhancing the Long-Range Dependence |
| title_full_unstemmed | Improving the Accuracy of Lane Detection by Enhancing the Long-Range Dependence |
| title_short | Improving the Accuracy of Lane Detection by Enhancing the Long-Range Dependence |
| title_sort | improving the accuracy of lane detection by enhancing the long range dependence |
| topic | lane detection long-range dependence CLRNet nonlocal mechanism feature pyramid network |
| url | https://www.mdpi.com/2079-9292/12/11/2518 |
| work_keys_str_mv | AT boliu improvingtheaccuracyoflanedetectionbyenhancingthelongrangedependence AT lifeng improvingtheaccuracyoflanedetectionbyenhancingthelongrangedependence AT qinglinzhao improvingtheaccuracyoflanedetectionbyenhancingthelongrangedependence AT guanghuili improvingtheaccuracyoflanedetectionbyenhancingthelongrangedependence AT yufengchen improvingtheaccuracyoflanedetectionbyenhancingthelongrangedependence |