1.College of Data Science and Application, Inner Mongolia University of Technology, Hohhot 010080, China
2.College of Information Engineering, Inner Mongolia University of Technology, Hohhot 010080, China
Show less
文章历史+
Received
Accepted
Published
2024-03-25
Issue Date
2026-02-27
PDF (5882K)
摘要
为提高城市快速路网拥堵识别的精确性,提出一种城市快速路网拥堵检测算法(congestion recognition mechanism in urban expressway,CRMU)。CRMU算法由特征提取、交通流指标计算和拥堵检测3个阶段组成。在特征提取阶段,为提升检测精度,采用K-means++算法对初始锚框值进行优化,结合BiFPN结构改进颈部网络,以提高检测精度和特征提取能力,利用提取的特征计算车速、车流量、车道数和车道长度等交通流特征值。在拥堵检测阶段,将提取的特征值输入改进核函数的支持向量机(support vector machine,SVM)模型,最终输出拥堵分类情况。实验结果表明,CRMU拥堵检测精度达到93.82%,优于传统拥堵检测算法,在城市快速路网拥堵检测方面更具优势。
Abstract
To improve the accuracy of urban expressway congestion recognition, a congestion detection algorithm (congestion recognition mechanism in urban expressway, CRMU) was proposed for congestion detection in urban expressway. CRMU was composed of three parts: feature extraction, traffic flow indicator calculation, and congestion detection. In the feature extraction stage, to enhance detection accuracy, the K-means++ algorithm was employed to optimize the initial anchor box values, combined with the BiFPN structure to improve the neck network, thereby enhancing detection accuracy and feature extraction capability. Subsequently, the extracted features were utilized to compute traffic flow indicators such as vehicle speed, traffic flow, number of lanes, and lane length. In the congestion detection stage, the extracted features were inputted into an improved kernel function-based support vector machine (SVM) model to ultimately output congestion classification. The experimental results verify that the detection accuracy of CRMU can reach 93.82%, surpassing traditional congestion detection algorithms, thus presenting superior performance in urban expressway congestion detection.
BIEM, LIUY, LIG, et al. Real-time vehicle detection algorithm based on a lightweight you-only-look-once(YOLOv5n-L) approach[J]. Expert Systems with Applications, 2023, 213: 119108.
RJOUBG, WAHABO A, BENTAHARJ, et al. Improving autonomous vehicles safety in snow weather using federated YOLO CNN learning[C]//International Conference on Mobile Web and Intelligent Information Systems, Cham: Springer International Publishing, 2021: 121-134.
[5]
ZHAOJ Y, HAOS N, DAIC X, et al. Improved vision-based vehicle detection and classification by optimized YOLOv4[J]. IEEE Access, 2022, 10: 8590-8603.
[6]
ZHUX K, LYU S, WANGX, et al. TPH-YOLOv5: improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios[C]//2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW). Montreal, BC, Canada: IEEE, 2021: 2778-2788.
[7]
LIUZ G, GAOY, DUQ Q, et al. YOLO-extract: improved YOLOv5 for aircraft object detection in remote sensing images[J]. IEEE Access, 2023, 11: 1742-1751.
[8]
XIANGX Z, WANGZ Y, QIAOY L. An improved YOLOv5 crack detection method combined with transformer[J]. IEEE Sensors Journal, 2022, 22(14): 14328-14335.
[9]
RENJ, WANGZ, ZHANGY, et al. YOLOv5-R: lightweight real-time detection based on improved YOLOv5[J]. Journal of Electronic Imaging, 2022, 31(3): 033033.
QUZ, GAOL Y, WANGS Y, et al. An improved YOLOv5 method for large objects detection with multi-scale feature cross-layer fusion network[J]. Image and Vision Computing, 2022, 125: 104518.
[12]
DANGT P, TRANN T, TOV H, et al. Improved YOLOv5 for real-time traffic signs recognition in bad weather conditions[J]. The Journal of Supercomputing, 2023, 79(10): 10706-10724.
[13]
HOUQ Z, LENGJ Q, MAG S, et al. An adaptive hybrid model for short-term urban traffic flow prediction[J]. Physica A: Statistical Mechanics and its Applications, 2019, 527: 121065.
[14]
ANJANEYULUM, KUBENDIRANM. Short-term traffic congestion prediction using hybrid deep learning technique[J]. Sustainability, 2022, 15(1): 74.
[15]
XUH B, JIANGC S. Deep belief network-based support vector regression method for traffic flow forecasting[J]. Neural Computing and Applications, 2020, 32(7): 2027-2036.
OLORUNSHOLAO E, IRHEBHUDEM E, EVWIEKPAEFEA E. A comparative study of YOLOv5 and YOLOv7 object detection algorithms[J]. Journal of Computing and Social Informatics, 2023, 2(1): 1-12.
[18]
CHENJ, MAIH S, LUOL B, et al. Effective feature fusion network in BIFPN for small object detection[C]//2021 IEEE International Conference on Image Processing (ICIP). Anchorage, AK, USA: IEEE, 2021: 699-703.
[19]
SONGD, THARMARASAR, ZHOUG J, et al. Multi-vehicle tracking using microscopic traffic models[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(1): 149-161.
[20]
KASHEFR. A boosted SVM classifier trained by incremental learning and decremental unlearning approach[J]. Expert Systems with Applications, 2021, 167: 114154.
[21]
MAIX C, ZHANGH, JIAX, et al. Faster R-CNN with classifier fusion for automatic detection of small fruits[J]. IEEE Transactions on Automation Science and Engineering, 2020, 17(3): 1555-1569.
[22]
TAHERI TAJARA, RAMAZANIA, MANSOORIZADEHM. A lightweight tiny-YOLOv3 vehicle detection approach[J]. Journal of Real-Time Image Processing, 2021, 18(6): 2389-2401.