Vehicle Detection in Congested Traffic Based on Simplified Weighted Dual-Path Feature Pyramid Network With Guided Anchoring

被引:5
|
作者
Luo, Jingqing [1 ]
Fang, Husheng [1 ]
Shao, Faming [1 ]
Hu, Cong [1 ]
Meng, Fanjie [2 ]
机构
[1] Army Engn Univ PLA, Coll Field Engn, Dept Mech Engn, Nanjing 210007, Peoples R China
[2] Univ PLA, Dept Test & Launch Aerosp Engn, Nanjing 210007, Peoples R China
基金
中国国家自然科学基金;
关键词
Feature extraction; Vehicle detection; Object detection; Robustness; Roads; Monitoring; Deep learning; Simplified weighted dual-path feature pyramid network; guided anchoring; DIoU-soft NMS; multi-task loss; OBJECT DETECTION;
D O I
10.1109/ACCESS.2021.3069216
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In modern life, traffic congestion is widespread in large and medium-sized cities in various countries. Multi-scale vehicle targets are densely distributed and occluded from each other in the images of crowded scenes. Vehicle detection in such scenarios is of great significance to urban traffic control, safety management and criminal investigation, but also has great challenges. Facing the special application in congested traffic, we propose Simplified Weighted Dual-path Network with Guided Anchoring framework to realize real-time vehicle detection. Firstly, a simplified weighted Dual-path Feature Pyramid Network (SWD-FPN) is used to improve the robustness of the model for multi-scale and partially occluded objects. Secondly, in order to improve the detection capability for vehicles with wide range of scale changes, the Guided Anchoring (GA) is applied to generate anchors of corresponding positions and scales according to the feature maps. Finally, for the challenge of vehicles intensive distribution, DIoU-soft NMS post-processing mechanism is introduced to reduce the missing alarm. Considering the class imbalance of vehicle detection in real-time traffic scenarios and the above improvements, multi-task loss is proposed for training. Ablation experiments are performed on UA-DETRAC dataset to further analyze the effect of different strategies on performance improvement. In addition, comparisons experiments on UA-DETRAC dataset and handcrafted Vehicles of Traffic (VOF) dataset are conducted to demonstrate the superiority of the proposed method over other state-of-the-art methods.
引用
收藏
页码:53219 / 53231
页数:13
相关论文
共 39 条
  • [1] Discriminative Semantic Feature Pyramid Network with Guided Anchoring for Logo Detection
    Zhang, Baisong
    Hou, Sujuan
    Karim, Awudu
    Wang, Jing
    Jia, Weikuan
    Zheng, Yuanjie
    MATHEMATICS, 2023, 11 (02)
  • [2] A scene text detection based on dual-path feature fusion
    Zhao P.
    Xu B.-P.
    Yan S.
    Liu Z.-Y.
    Kongzhi yu Juece/Control and Decision, 2021, 36 (09): : 2179 - 2186
  • [3] Rail Surface Defect Detection Based on Dual-Path Feature Fusion
    Zhong, Yinfeng
    Chen, Guorong
    ELECTRONICS, 2024, 13 (13)
  • [4] DP-FishNet: Dual-path Pyramid Vision Transformer-based underwater fish detection network
    Liu, Yang
    An, Dong
    Ren, Yinjie
    Zhao, Jian
    Zhang, Chi
    Cheng, Jiahui
    Liu, Jincun
    Wei, Yaoguang
    EXPERT SYSTEMS WITH APPLICATIONS, 2024, 238
  • [5] Dual-Path Reconstruction Guided Segmentation Network for Unsupervised Anomaly Detection and Localization
    Xiao, Junwei
    Deng, Lei
    Chen, Zhixiang
    Li, Xiu
    Chen, Baohua
    Yin, Hanxi
    2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN, 2023,
  • [6] Dual-path multi-branch feature residual network for salient object detection
    Chen, Zhensen
    Lu, Yaosheng
    Long, Shun
    Bai, Jieyun
    ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2024, 133
  • [7] Dual-path feature extraction based hybrid intrusion detection in IoT networks
    Silivery, Arun Kumar
    Rao, Kovvur Ram Mohan
    Solleti, Ramana
    COMPUTERS & ELECTRICAL ENGINEERING, 2025, 122
  • [8] In-field cotton detection algorithm based on the dual-path feature extraction
    Xu, Yang
    Li, Yanan
    Wu, Hao
    Wen, Hongyu
    JOURNAL OF ELECTRONIC IMAGING, 2021, 30 (04)
  • [9] Automatic detection method for tobacco beetles combining multi-scale global residual feature pyramid network and dual-path deformable attention
    Chen, Yuling
    Li, Xiaoxia
    Lv, Nianzu
    He, Zhenxiang
    Wu, Bin
    SCIENTIFIC REPORTS, 2024, 14 (01)
  • [10] Automatic detection method for tobacco beetles combining multi-scale global residual feature pyramid network and dual-path deformable attention
    Yuling Chen
    Xiaoxia Li
    Nianzu Lv
    Zhenxiang He
    Bin Wu
    Scientific Reports, 14