Affiliation:
1. Wuhan University of Technology
Abstract
<div class="section abstract"><div class="htmlview paragraph">As a key technology of intelligent transportation system, vehicle type recognition plays an important role in ensuring traffic safety,optimizing traffic management and improving traffic efficiency, which provides strong support for the development of modern society and the intelligent construction of traffic system. Aiming at the problems of large number of parameters, low detection efficiency and poor real-time performance in existing vehicle type recognition algorithms, this paper proposes an improved vehicle type recognition algorithm based on YOLOv5. Firstly, the lightweight network model MobileNet-V3 is used to replace the backbone feature extraction network CSPDarknet53 of the YOLOv5 model. The parameter quantity and computational complexity of the model are greatly reduced by replacing the standard convolution with the depthwise separable convolution, and enabled the model to maintain higher accuracy while having faster reasoning speed. Secondly, the attention mechanism in MobileNet-V3 is improved, and a more efficient coordinate attention module is embedded to enhance the model ‘s attention to key features, so as to further improve the accuracy of vehicle type recognition. Finally, in order to better integrate different levels of features, the weighted bidirectional feature pyramid network BiFPN structure is introduced. By processing the top-down and bottom-up paths at the same time, the feature maps of different resolutions are fused, so that the model can obtain richer feature information at multiple scales, thereby improving detection efficiency and accuracy. The experimental results show that the improved algorithm has a vehicle type recognition accuracy of 96% on the BIT-Vehicle dataset. Compared with the original YOLOv5 model, the number of model parameters is reduced by 47%, and the detection speed is increased by 54.6%. Therefore, the improved algorithm in this paper can better meet the requirements of small parameter quantity and high detection efficiency, and is suitable for real-time vehicle type recognition.</div></div>
Reference20 articles.
1. Shengdong , M. ,
Zhengxian , X. , and
Yixiang , T.
Intelligent Traffic Control System Based on Cloud Computing and Big Data Mining IEEE Transactions on Industrial Informatics 99 2019 1 1 10.1109/TII.2019.2929060
2. Meher , S.K. and
Murty , M.N.
Efficient Method of Moving Shadow Detection and Vehicle Classification AEU: Archiv fur Elektronik und Ubertragungstechnik: Electronic and Communication 8 2013 67
3. Hsieh , J.W.
et al.
Vehicle Make and Model Recognition Using Symmetrical SURF Advanced Video and Signal Based Surveillance (AVSS), 2013 10th IEEE International Conference on IEEE 2013 10.1109/AVSS.2013.6636685
4. Zhuoyuan , H. ,
Cao Yudong , H. , and
Qiang.
Vehicle Type Recognition Algorithm Based on Fusion Features Journal of Liaoning University of Technology (Natural Science Edition) 6 2016 361 363
5. Ren , S.
et al.
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks IEEE Transactions on Pattern Analysis & Machine Intelligence 39 6 2017 1137 1149 10.1109/TPAMI.2016.2577031