A Multi-Scale Traffic Object Detection Algorithm for Road Scenes Based on Improved YOLOv5

被引:18
|
作者
Li, Ang [1 ]
Sun, Shijie [1 ]
Zhang, Zhaoyang [1 ]
Feng, Mingtao [2 ,3 ]
Wu, Chengzhong [3 ]
Li, Wang [4 ]
机构
[1] Changan Univ, Sch Informat Engn, Xian 710064, Peoples R China
[2] Xidian Univ, Sch Comp Sci & Technol, Xian 710000, Peoples R China
[3] Hunan Univ, Natl Engn Lab Robot Visual Percept & Control Techn, Changsha 410000, Peoples R China
[4] CRRC Zhuzhou Elect Locomot Co Ltd, Zhuzhou 412000, Peoples R China
基金
中国国家自然科学基金;
关键词
road scenes; object detection; YOLOv5; multi-scale; attention mechanism;
D O I
10.3390/electronics12040878
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Object detection in road scenes is a task that has recently become popular and it is also an important part of intelligent transportation systems. Due to the different locations of cameras in the road scenes, the size of the traffic objects captured varies greatly, which imposes a burden on the network optimization. In addition, in some dense traffic scenes, the size of the traffic objects captured is extremely small and it is easy to miss detection and to encounter false detection. In this paper, we propose an improved multi-scale YOLOv5s algorithm based on the YOLOv5s algorithm. In detail, we add a detection head for extremely small objects to the original YOLOv5s model, which significantly improves the accuracy in detecting extremely small traffic objects. A content-aware reassembly of features (CARAFE) module is introduced in the feature fusion part to enhance the feature fusion. A new SPD-Conv CNN Module is introduced instead of the original convolutional structure to enhance the overall computational efficiency of the model. Finally, the normalization-based attention module (NAM) is introduced, allowing the model to focus on more useful information during training and significantly improving detection accuracy. The experimental results demonstrate that compared with the original YOLOv5s algorithm, the detection accuracy of the multi-scale YOLOv5s model proposed in this paper is improved by 7.1% on the constructed diverse traffic scene datasets. The improved multi-scale YOLOv5s algorithm also maintains the highest detection accuracy among the current mainstream object detection algorithms and is superior in accomplishing the task of detecting traffic objects in complex road scenes.
引用
收藏
页数:16
相关论文
共 50 条
  • [1] Research on pedestrian object detection algorithm in urban road scenes based on improved YOLOv5
    Liu, Zhaohui
    Wang, Xiao
    [J]. Journal of Intelligent and Fuzzy Systems, 2024, 1 (01):
  • [2] ETSR-YOLO: An improved multi-scale traffic sign detection algorithm based on YOLOv5
    Liu, Haibin
    Zhou, Kui
    Zhang, Youbing
    Zhang, Yufeng
    [J]. PLOS ONE, 2023, 18 (12):
  • [3] Road object detection method based on improved YOLOv5 algorithm
    Wang, Hong-Zhi
    Song, Ming-Xuan
    Cheng, Chao
    Xie, Dong-Xuan
    [J]. Jilin Daxue Xuebao (Gongxueban)/Journal of Jilin University (Engineering and Technology Edition), 2024, 54 (09): : 2658 - 2667
  • [4] Traffic Target Detection in Complex Scenes Based on Improved YOLOv5 Algorithm
    Gu, De-Ying
    Luo, Yu-Lun
    Li, Wen-Chao
    [J]. Dongbei Daxue Xuebao/Journal of Northeastern University, 2022, 43 (08): : 1073 - 1079
  • [5] Improved YOLOv5 Small Object Detection Algorithm in Moving Scenes
    Zhu, Ruixin
    Yang, Fuxing
    [J]. Computer Engineering and Applications, 2023, 59 (10): : 196 - 203
  • [6] A multi-scale cucumber disease detection method in natural scenes based on YOLOv5
    Li, Shufei
    Li, Kaiyu
    Qiao, Yan
    Zhang, Lingxian
    [J]. COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2022, 202
  • [7] A Kind of Water Surface Multi-Scale Object Detection Method Based on Improved YOLOv5 Network
    Ma, Zhongli
    Wan, Yi
    Liu, Jiajia
    An, Ruojin
    Wu, Lili
    [J]. MATHEMATICS, 2023, 11 (13)
  • [8] Object detection and localization algorithm in agricultural scenes based on YOLOv5
    Yang, Jiachen
    Han, Mengqi
    He, Jingyi
    Wen, Jiabao
    Chen, Desheng
    Wang, Yibo
    [J]. JOURNAL OF ELECTRONIC IMAGING, 2023, 32 (05) : 52402
  • [9] Multi-scale Forest Flame Detection Based on Improved and Optimized YOLOv5
    Zhou, Mengdong
    Liu, Shuai
    Li, Jianjun
    [J]. FIRE TECHNOLOGY, 2023, 59 (06) : 3689 - 3708
  • [10] Multi-scale Forest Flame Detection Based on Improved and Optimized YOLOv5
    Mengdong Zhou
    Shuai Liu
    Jianjun Li
    [J]. Fire Technology, 2023, 59 : 3689 - 3708