A Novel Method for Unexpected Obstacle Detection in the Traffic Environment Based on Computer Vision
Obstacle detection is the basis for the Advanced Driving Assistance System (ADAS) to take obstacle avoidance measures. However, it is a very essential and challenging task to detect unexpected obstacles on the road. To this end, an unexpected obstacle detection method based on computer vision is pro...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/18/8937 |
| _version_ | 1797491643360739328 |
|---|---|
| author | Wenyan Ci Tianxiang Xu Runze Lin Shan Lu |
| author_facet | Wenyan Ci Tianxiang Xu Runze Lin Shan Lu |
| author_sort | Wenyan Ci |
| collection | DOAJ |
| description | Obstacle detection is the basis for the Advanced Driving Assistance System (ADAS) to take obstacle avoidance measures. However, it is a very essential and challenging task to detect unexpected obstacles on the road. To this end, an unexpected obstacle detection method based on computer vision is proposed. We first present two independent methods for the detection of unexpected obstacles: a semantic segmentation method that can highlight the contextual information of unexpected obstacles on the road and an open-set recognition algorithm that can distinguish known and unknown classes according to the uncertainty degree. Then, the detection results of the two methods are input into the Bayesian framework in the form of probabilities for the final decision. Since there is a big difference between semantic and uncertainty information, the fusion results reflect the respective advantages of the two methods. The proposed method is tested on the Lost and Found dataset and evaluated by comparing it with the various obstacle detection methods and fusion strategies. The results show that our method improves the detection rate while maintaining a relatively low false-positive rate. Especially when detecting unexpected long-distance obstacles, the fusion method outperforms the independent methods and keeps a high detection rate. |
| first_indexed | 2024-03-10T00:52:15Z |
| format | Article |
| id | doaj.art-cc955c08994a411c9ada136ad81a6f5a |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T00:52:15Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-cc955c08994a411c9ada136ad81a6f5a2023-11-23T14:50:27ZengMDPI AGApplied Sciences2076-34172022-09-011218893710.3390/app12188937A Novel Method for Unexpected Obstacle Detection in the Traffic Environment Based on Computer VisionWenyan Ci0Tianxiang Xu1Runze Lin2Shan Lu3School of Engineering, Huzhou University, Huzhou 313000, ChinaSchool of Engineering, Huzhou University, Huzhou 313000, ChinaState Key Laboratory of Industrial Control Technology, Zhejiang University, Hangzhou 310027, ChinaInstitute of Intelligence Science and Engineering, Shenzhen Polytechnic, Shenzhen 518055, ChinaObstacle detection is the basis for the Advanced Driving Assistance System (ADAS) to take obstacle avoidance measures. However, it is a very essential and challenging task to detect unexpected obstacles on the road. To this end, an unexpected obstacle detection method based on computer vision is proposed. We first present two independent methods for the detection of unexpected obstacles: a semantic segmentation method that can highlight the contextual information of unexpected obstacles on the road and an open-set recognition algorithm that can distinguish known and unknown classes according to the uncertainty degree. Then, the detection results of the two methods are input into the Bayesian framework in the form of probabilities for the final decision. Since there is a big difference between semantic and uncertainty information, the fusion results reflect the respective advantages of the two methods. The proposed method is tested on the Lost and Found dataset and evaluated by comparing it with the various obstacle detection methods and fusion strategies. The results show that our method improves the detection rate while maintaining a relatively low false-positive rate. Especially when detecting unexpected long-distance obstacles, the fusion method outperforms the independent methods and keeps a high detection rate.https://www.mdpi.com/2076-3417/12/18/8937unexpected obstacle detectioncomputer visionsemantic segmentationopen-set recognition algorithmuncertainty degreeBayesian fusion |
| spellingShingle | Wenyan Ci Tianxiang Xu Runze Lin Shan Lu A Novel Method for Unexpected Obstacle Detection in the Traffic Environment Based on Computer Vision Applied Sciences unexpected obstacle detection computer vision semantic segmentation open-set recognition algorithm uncertainty degree Bayesian fusion |
| title | A Novel Method for Unexpected Obstacle Detection in the Traffic Environment Based on Computer Vision |
| title_full | A Novel Method for Unexpected Obstacle Detection in the Traffic Environment Based on Computer Vision |
| title_fullStr | A Novel Method for Unexpected Obstacle Detection in the Traffic Environment Based on Computer Vision |
| title_full_unstemmed | A Novel Method for Unexpected Obstacle Detection in the Traffic Environment Based on Computer Vision |
| title_short | A Novel Method for Unexpected Obstacle Detection in the Traffic Environment Based on Computer Vision |
| title_sort | novel method for unexpected obstacle detection in the traffic environment based on computer vision |
| topic | unexpected obstacle detection computer vision semantic segmentation open-set recognition algorithm uncertainty degree Bayesian fusion |
| url | https://www.mdpi.com/2076-3417/12/18/8937 |
| work_keys_str_mv | AT wenyanci anovelmethodforunexpectedobstacledetectioninthetrafficenvironmentbasedoncomputervision AT tianxiangxu anovelmethodforunexpectedobstacledetectioninthetrafficenvironmentbasedoncomputervision AT runzelin anovelmethodforunexpectedobstacledetectioninthetrafficenvironmentbasedoncomputervision AT shanlu anovelmethodforunexpectedobstacledetectioninthetrafficenvironmentbasedoncomputervision AT wenyanci novelmethodforunexpectedobstacledetectioninthetrafficenvironmentbasedoncomputervision AT tianxiangxu novelmethodforunexpectedobstacledetectioninthetrafficenvironmentbasedoncomputervision AT runzelin novelmethodforunexpectedobstacledetectioninthetrafficenvironmentbasedoncomputervision AT shanlu novelmethodforunexpectedobstacledetectioninthetrafficenvironmentbasedoncomputervision |