A Multi-Stream Approach to Mixed-Traffic Accident Recognition Using Deep Learning
Road traffic accidents are a leading cause of injuries and fatalities globally, prompting extensive research into deep learning-based accident recognition models for their superior performance in computer vision tasks. However, most studies focus on non-mixed traffic environments, where detection is...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10781397/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846123663862005760 |
|---|---|
| author | Swee Tee Fu Lau Bee Theng Brian Loh Chung Shiong Chris McCarthy Mark Tee Kit Tsun |
| author_facet | Swee Tee Fu Lau Bee Theng Brian Loh Chung Shiong Chris McCarthy Mark Tee Kit Tsun |
| author_sort | Swee Tee Fu |
| collection | DOAJ |
| description | Road traffic accidents are a leading cause of injuries and fatalities globally, prompting extensive research into deep learning-based accident recognition models for their superior performance in computer vision tasks. However, most studies focus on non-mixed traffic environments, where detection is simpler due to predictable traffic patterns and uniform vehicle types. In contrast, mixed-traffic scenarios present greater challenges as diverse vehicles, motorcyclists, and pedestrians move unpredictably. Models relying on a single type of perception are effective in structured traffic but struggle to handle the complexities of mixed-traffic environments. This study proposes a novel multi-stream deep learning model called Accident Recognition in Mixed-Traffic Scene (ARMS), which integrates three distinct streams: the first stream analyzes the overall accident scene, the second focuses on mixed-traffic accident features, and the third examines vehicle motion abnormalities through object detection and tracking, aimed at improving road accident recognition accuracy in mixed-traffic environments at intersections. This model is trained and evaluated using datasets from CADP, UA-DETRAC, and supplementary online sources. The results demonstrate that the ARMS model achieves an accuracy of 93.3%, with performance improving significantly through the fusion of the individual streams. Additionally, the ARMS model was evaluated using two publicly available standard datasets, which further highlights its improved performance in recognizing mixed-traffic accidents compared to existing studies. |
| format | Article |
| id | doaj-art-dc5d871b78f441ccb5b489c84ec4adb9 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-dc5d871b78f441ccb5b489c84ec4adb92024-12-14T00:00:56ZengIEEEIEEE Access2169-35362024-01-011218523218524910.1109/ACCESS.2024.351279410781397A Multi-Stream Approach to Mixed-Traffic Accident Recognition Using Deep LearningSwee Tee Fu0https://orcid.org/0000-0002-9813-0625Lau Bee Theng1https://orcid.org/0000-0002-6071-6399Brian Loh Chung Shiong2https://orcid.org/0000-0002-4034-4233Chris McCarthy3https://orcid.org/0000-0003-3848-1631Mark Tee Kit Tsun4https://orcid.org/0000-0002-4413-2000Faculty of Engineering, Computing and Science, Swinburne University of Technology Sarawak Campus, Kuching, Sarawak, MalaysiaFaculty of Engineering, Computing and Science, Swinburne University of Technology Sarawak Campus, Kuching, Sarawak, MalaysiaFaculty of Engineering, Computing and Science, Swinburne University of Technology Sarawak Campus, Kuching, Sarawak, MalaysiaSchool of Science, Computing and Engineering Technologies (SoSCET), Swinburne University of Technology, Melbourne Campus, Hawthorn, VIC, AustraliaFaculty of Engineering, Computing and Science, Swinburne University of Technology Sarawak Campus, Kuching, Sarawak, MalaysiaRoad traffic accidents are a leading cause of injuries and fatalities globally, prompting extensive research into deep learning-based accident recognition models for their superior performance in computer vision tasks. However, most studies focus on non-mixed traffic environments, where detection is simpler due to predictable traffic patterns and uniform vehicle types. In contrast, mixed-traffic scenarios present greater challenges as diverse vehicles, motorcyclists, and pedestrians move unpredictably. Models relying on a single type of perception are effective in structured traffic but struggle to handle the complexities of mixed-traffic environments. This study proposes a novel multi-stream deep learning model called Accident Recognition in Mixed-Traffic Scene (ARMS), which integrates three distinct streams: the first stream analyzes the overall accident scene, the second focuses on mixed-traffic accident features, and the third examines vehicle motion abnormalities through object detection and tracking, aimed at improving road accident recognition accuracy in mixed-traffic environments at intersections. This model is trained and evaluated using datasets from CADP, UA-DETRAC, and supplementary online sources. The results demonstrate that the ARMS model achieves an accuracy of 93.3%, with performance improving significantly through the fusion of the individual streams. Additionally, the ARMS model was evaluated using two publicly available standard datasets, which further highlights its improved performance in recognizing mixed-traffic accidents compared to existing studies.https://ieeexplore.ieee.org/document/10781397/Deep learningimage classificationobject detectionobject trackingroad accidents |
| spellingShingle | Swee Tee Fu Lau Bee Theng Brian Loh Chung Shiong Chris McCarthy Mark Tee Kit Tsun A Multi-Stream Approach to Mixed-Traffic Accident Recognition Using Deep Learning IEEE Access Deep learning image classification object detection object tracking road accidents |
| title | A Multi-Stream Approach to Mixed-Traffic Accident Recognition Using Deep Learning |
| title_full | A Multi-Stream Approach to Mixed-Traffic Accident Recognition Using Deep Learning |
| title_fullStr | A Multi-Stream Approach to Mixed-Traffic Accident Recognition Using Deep Learning |
| title_full_unstemmed | A Multi-Stream Approach to Mixed-Traffic Accident Recognition Using Deep Learning |
| title_short | A Multi-Stream Approach to Mixed-Traffic Accident Recognition Using Deep Learning |
| title_sort | multi stream approach to mixed traffic accident recognition using deep learning |
| topic | Deep learning image classification object detection object tracking road accidents |
| url | https://ieeexplore.ieee.org/document/10781397/ |
| work_keys_str_mv | AT sweeteefu amultistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning AT laubeetheng amultistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning AT brianlohchungshiong amultistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning AT chrismccarthy amultistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning AT markteekittsun amultistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning AT sweeteefu multistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning AT laubeetheng multistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning AT brianlohchungshiong multistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning AT chrismccarthy multistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning AT markteekittsun multistreamapproachtomixedtrafficaccidentrecognitionusingdeeplearning |