A Multi-Stream Approach to Mixed-Traffic Accident Recognition Using Deep Learning

被引：0
|
作者：
Fu, Swee Tee [1 ]
Theng, Lau Bee [1 ]
Shiong, Brian Loh Chung [1 ]
Mccarthy, Chris [2 ]
Tsun, Mark Tee Kit [1 ]
机构：
[1] Swinburne University of Technology, Faculty of Engineering, Computing and Science, Sarawak,Sarawak Campus, Kuching,93350, Malaysia
[2] Swinburne University of Technology, School of Science, Computing and Engineering Technologies (SoSCET), Melbourne Campus, Hawthorn,VIC,3122, Australia
来源：
IEEE Access | 2024年 / 12卷
关键词：
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. © 2013 IEEE;
D O I：
10.1109/ACCESS.2024.3512794
中图分类号：
学科分类号：
摘要：
引用
收藏
页码：185232 / 185249
相关论文
共 50 条
  • [41] Multi-Stream Interaction Networks for Human Action Recognition
    Wang, Haoran
    Yu, Baosheng
    Li, Jiaqi
    Zhang, Linlin
    Chen, Dongyue
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2022, 32 (05) : 3050 - 3060
  • [42] Multi-stream Recognition of Noisy Speech with Performance Monitoring
    Variani, Ehsan
    Li, Feipeng
    Hermansky, Hynek
    14TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2013), VOLS 1-5, 2013, : 2977 - 2980
  • [43] Multi-Stream End-to-End Speech Recognition
    Li, Ruizhi
    Wang, Xiaofei
    Mallidi, Sri Harish
    Watanabe, Shinji
    Hori, Takaaki
    Hermansky, Hynek
    IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2020, 28 (646-655) : 646 - 655
  • [44] Multi-stream face recognition for crime-fightin
    Jassim, Sabah A.
    Sellahewa, Harin
    BIOMETRIC TECHNOLOGY FOR HUMAN IDENTIFICATION IV, 2007, 6539
  • [45] An Independent Trajectory Advisory System in a Mixed-Traffic Condition: A Reinforcement Learning-Based Approach
    Rostami-Shahrbabaki, Majid
    Niels, Tanja
    Hamzehi, Sascha
    Bogenberger, Klaus
    IFAC PAPERSONLINE, 2020, 53 (02): : 15667 - 15673
  • [46] A spatiotemporal multi-stream learning framework based on attention mechanism for automatic modulation recognition
    Wang, Xu
    Liu, Dejun
    Zhang, Yuhao
    Li, Yang
    Wu, Shiwei
    DIGITAL SIGNAL PROCESSING, 2022, 130
  • [47] Sign Language Recognition Based on Position and Movement Using Multi-Stream HMM
    Maebatake, Masaru
    Suzuki, Iori
    Nishida, Masafumi
    Horiuchi, Yasuo
    Kuroiwa, Shingo
    PROCEEDINGS OF THE SECOND INTERNATIONAL SYMPOSIUM ON UNIVERSAL COMMUNICATION, 2008, : 478 - 481
  • [48] Arabic and Latin handwritting recognition using multi-stream hidden Markov models
    Kessentini, Yousri
    Paquet, Thierry
    Ben Hamadou, AbdelMajid
    TRAITEMENT DU SIGNAL, 2009, 26 (05) : 395 - 407
  • [49] A spatiotemporal multi-stream learning framework based on attention mechanism for automatic modulation recognition
    College of Information Science and Engineering, China University of Petroleum-Beijing, Beijing, China
    Digital Signal Process Rev J,
  • [50] New modeling approach for mixed-traffic streams with nonmotorized vehicles
    Oketch, TG
    PEDESTRIAN AND BICYCLE TRANSPORTATION RESEARCH 2000: SAFETY AND HUMAN PERFORMANCE, 2000, (1705): : 61 - 69
12345