Real‐time traffic cone detection for autonomous driving based on YOLOv4
Abstract A temporary road composed of traffic cones is an indispensable practical scene for the realization of automatic driving technology. However, the detection of traffic cones is a challenging issue because of their small volume and unfixed position. This work proposes a novel method that fuses...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-10-01
|
| Series: | IET Intelligent Transport Systems |
| Online Access: | https://doi.org/10.1049/itr2.12212 |
| _version_ | 1818020479270649856 |
|---|---|
| author | Qinghua Su Haodong Wang Min Xie Yue Song Shaobo Ma Boxiong Li Ying Yang Liyong Wang |
| author_facet | Qinghua Su Haodong Wang Min Xie Yue Song Shaobo Ma Boxiong Li Ying Yang Liyong Wang |
| author_sort | Qinghua Su |
| collection | DOAJ |
| description | Abstract A temporary road composed of traffic cones is an indispensable practical scene for the realization of automatic driving technology. However, the detection of traffic cones is a challenging issue because of their small volume and unfixed position. This work proposes a novel method that fuses colour and depth image information for traffic cone detection. Traffic cones are captured by a special machine vision system consisting of two monochrome cameras for cone distance perception and two colour cameras for cone colour acquisition. Via the YOLOv4 algorithm based on the Darknet platform and a detection result matching algorithm, the position of the traffic cone can be obtained and path planning can be performed. The results of experiments show that the proposed method can recognize red, blue, and yellow traffic cones in colour images with an average detection time of 35.46 ms and respective accuracies of 97.51%, 98.63%, and 97.29%. Compared with the previous traffic cone detection research, the proposed algorithm was found to exhibit advantages in small target sensitivity and overall detection accuracy in both static and dynamic experiments. |
| first_indexed | 2024-04-14T08:06:45Z |
| format | Article |
| id | doaj.art-f7d8b825433c4577be4840564211860b |
| institution | Directory Open Access Journal |
| issn | 1751-956X 1751-9578 |
| language | English |
| last_indexed | 2024-04-14T08:06:45Z |
| publishDate | 2022-10-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Intelligent Transport Systems |
| spelling | doaj.art-f7d8b825433c4577be4840564211860b2022-12-22T02:04:44ZengWileyIET Intelligent Transport Systems1751-956X1751-95782022-10-0116101380139010.1049/itr2.12212Real‐time traffic cone detection for autonomous driving based on YOLOv4Qinghua Su0Haodong Wang1Min Xie2Yue Song3Shaobo Ma4Boxiong Li5Ying Yang6Liyong Wang7Key Laboratory of Modern Measurement and Control Technology Ministry of Education Beijing Information Science and Technology University Beijing ChinaKey Laboratory of Modern Measurement and Control Technology Ministry of Education Beijing Information Science and Technology University Beijing ChinaKey Laboratory of Modern Measurement and Control Technology Ministry of Education Beijing Information Science and Technology University Beijing ChinaKey Laboratory of Modern Measurement and Control Technology Ministry of Education Beijing Information Science and Technology University Beijing ChinaKey Laboratory of Modern Measurement and Control Technology Ministry of Education Beijing Information Science and Technology University Beijing ChinaKey Laboratory of Modern Measurement and Control Technology Ministry of Education Beijing Information Science and Technology University Beijing ChinaKey Laboratory of Modern Measurement and Control Technology Ministry of Education Beijing Information Science and Technology University Beijing ChinaKey Laboratory of Modern Measurement and Control Technology Ministry of Education Beijing Information Science and Technology University Beijing ChinaAbstract A temporary road composed of traffic cones is an indispensable practical scene for the realization of automatic driving technology. However, the detection of traffic cones is a challenging issue because of their small volume and unfixed position. This work proposes a novel method that fuses colour and depth image information for traffic cone detection. Traffic cones are captured by a special machine vision system consisting of two monochrome cameras for cone distance perception and two colour cameras for cone colour acquisition. Via the YOLOv4 algorithm based on the Darknet platform and a detection result matching algorithm, the position of the traffic cone can be obtained and path planning can be performed. The results of experiments show that the proposed method can recognize red, blue, and yellow traffic cones in colour images with an average detection time of 35.46 ms and respective accuracies of 97.51%, 98.63%, and 97.29%. Compared with the previous traffic cone detection research, the proposed algorithm was found to exhibit advantages in small target sensitivity and overall detection accuracy in both static and dynamic experiments.https://doi.org/10.1049/itr2.12212 |
| spellingShingle | Qinghua Su Haodong Wang Min Xie Yue Song Shaobo Ma Boxiong Li Ying Yang Liyong Wang Real‐time traffic cone detection for autonomous driving based on YOLOv4 IET Intelligent Transport Systems |
| title | Real‐time traffic cone detection for autonomous driving based on YOLOv4 |
| title_full | Real‐time traffic cone detection for autonomous driving based on YOLOv4 |
| title_fullStr | Real‐time traffic cone detection for autonomous driving based on YOLOv4 |
| title_full_unstemmed | Real‐time traffic cone detection for autonomous driving based on YOLOv4 |
| title_short | Real‐time traffic cone detection for autonomous driving based on YOLOv4 |
| title_sort | real time traffic cone detection for autonomous driving based on yolov4 |
| url | https://doi.org/10.1049/itr2.12212 |
| work_keys_str_mv | AT qinghuasu realtimetrafficconedetectionforautonomousdrivingbasedonyolov4 AT haodongwang realtimetrafficconedetectionforautonomousdrivingbasedonyolov4 AT minxie realtimetrafficconedetectionforautonomousdrivingbasedonyolov4 AT yuesong realtimetrafficconedetectionforautonomousdrivingbasedonyolov4 AT shaoboma realtimetrafficconedetectionforautonomousdrivingbasedonyolov4 AT boxiongli realtimetrafficconedetectionforautonomousdrivingbasedonyolov4 AT yingyang realtimetrafficconedetectionforautonomousdrivingbasedonyolov4 AT liyongwang realtimetrafficconedetectionforautonomousdrivingbasedonyolov4 |