Improved YOLOv7 Traffic Sign Detection Algorithm in Complex Scenarios

doi:10.19678/j.issn.1000-3428.0069132

Abstract

Abstract:

To solve the problems of misdetection and omission of traffic signs by existing target detection algorithms in complex scenarios, a traffic sign detection algorithm YOLOv7-MBFE is proposed to improve YOLOv7. First, the proposed multi-branch feature extraction module based on dilated convolution controls the shortest and longest gradient paths, thereby enhancing the feature extraction capability of the model. Second, an asymptotic feature pyramid network is built in the header to fully integrate feature information at different levels and improve the feature expression of the model. A channel attention mechanism is further introduced into the SPPCSPC module to adaptively adjust the weights of the channels, enhance the feature interaction between different channels, and integrate the multi-head self-attention mechanism into the downsampling stage, which enhances the ability of the model to perceive the global contextual information and improves its detection performance in complex scenarios. Finally, Focal-EIoU is used to replace the loss function in the original YOLOv7 model to make the model focus more on high-quality anchor frames, accelerate convergence speed, and improve robustness. In numerous experiments, the algorithm was applied to a Chinese traffic sign detection dataset. The experimental results show that the algorithm improved mean Average Precision (mAP) by 9.25%, accuracy by 3.92%, and recall by 5.19% compared with the YOLOv7 algorithm, significantly mitigating the problems of misdetection and missed detection in complex scenes and achieving better detection performance than those of the original algorithm and classical target detection algorithms.

Key words: target detection, traffic sign recognition, YOLOv7, multi-branch feature extraction, multi-scale feature fusion, attention mechanism

摘要：

为解决现有目标检测算法在复杂场景下对交通标志的误检、漏检等问题, 提出一种改进YOLOv7的交通标志检测算法YOLOv7-MBFE。首先, 提出一种基于膨胀卷积的多分支特征提取模块, 控制最短和最长的梯度路径, 增强模型的特征提取能力; 其次, 在头部网络中构建渐进特征金字塔结构, 充分融合不同层次的特征信息, 改善模型的特征表达能力; 在SPPCSPC模块中引入通道注意力机制, 自适应调整通道的权重, 增强不同通道之间的特征交互, 并将多头自注意力机制融合至下采样阶段, 增强模型对全局上下文信息的感知能力, 提高模型在复杂场景下的检测性能; 最后, 使用Focal-EIoU替换原YOLOv7模型中的损失函数, 使模型更专注于高质量的锚框, 加快模型的收敛速度, 提高模型的鲁棒性。在中国交通标志检测数据集上进行大量实验, 结果表明, 相较于YOLOv7算法, 该算法的平均精度均值(mAP)提升了9.25%, 准确率提升了3.92%, 召回率提升了5.19%。改进后的算法能够显著改善复杂场景下的误检、漏检等问题, 检测效果优于原始算法和经典目标检测算法。

关键词: 目标检测, 交通标志识别, YOLOv7, 多分支特征提取, 多尺度特征融合, 注意力机制

XU Ming, QU Taipeng, JIANG Yanji. Improved YOLOv7 Traffic Sign Detection Algorithm in Complex Scenarios[J]. Computer Engineering, 2025, 51(2): 335-343.

许明, 屈泰澎, 姜彦吉. 改进YOLOv7在复杂场景下的交通标志检测算法[J]. 计算机工程, 2025, 51(2): 335-343.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0069132

https://www.ecice06.com/EN/Y2025/V51/I2/335

Figures/Tables 9

Fig.1 Improved YOLOv7 network structure

Fig.2 Dilated convolution-based multi-branch feature extraction module structure

Fig.3 Squeeze-Excitement module structure

Fig.4 Multi-head self attention mechanism structure

Fig.5 Comparison of loss functions

Fig.6 Comparison of detection effects in different scenarios

References 30

1	DE LA ESCALERA A , MORENO L E , SALICHS M A , et al. Road traffic sign detection and classification. IEEE Transactions on Industrial Electronics, 1997, 44 (6): 848- 859. doi: 10.1109/41.649946
2	HOUBEN S, STALLKAMP J, SALMEN J, et al. Detection of traffic signs in real-world images: the German traffic sign detection benchmark[C]//Proceedings of the 2013 International Joint Conference on Neural Networks. Washington D.C., USA: IEEE Press, 2013: 1-8.
3	ZHU Z, LIANG D, ZHANG S H, et al. Traffic-sign detection and classification in the wild[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Washington D.C., USA: IEEE Press, 2016: 2110-2118.
4	YANG Y, LUO H L, XU H R, et al. Towards real-time traffic sign detection and classification[C]//Proceedings of the 17th International IEEE Conference on Intelligent Transportation Systems. Washington D.C., USA: IEEE Press, 2014: 2022-2031.
5	BOSER B E, GUYON I M, VAPNIK V N. A training algorithm for optimal margin classifiers[C]//Proceedings of the 5th Annual Workshop on Computational Learning Theory. New York, USA: ACM Press, 1992: 144-152.
6	DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington D.C., USA: IEEE Press, 2005: 886-893.
7	GIRSHICK R. Fast R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision. Washington D.C., USA: IEEE Press, 2015: 1440-1448.
8	REN S , HE K , GIRSHICK R , et al. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39 (6): 1137- 1149. doi: 10.1109/TPAMI.2016.2577031
9	HE K M, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision. Washington D.C., USA: IEEE Press, 2017: 2961-2969.
10	GONG C , LI A , SONG Y , et al. Traffic sign recognition based on the YOLOv3 algorithm. Sensors (Basel, Switzerland), 2022, 22 (23): 9345. doi: 10.3390/s22239345
11	REHMAN Y , AMANULLAH H , SAQIB BHATTI D M , et al. Detection of small size traffic signs using regressive anchor box selection and DBL layer tweaking in YOLOv3. Applied Sciences, 2021, 11 (23): 11555. doi: 10.3390/app112311555
12	彭成, 张乔虹, 唐朝晖, 等. 基于YOLOv5增强模型的口罩佩戴检测方法研究. 计算机工程, 2022, 48 (4): 39- 49. doi: 10.19678/j.issn.1000-3428.0061502
	PENG C , ZHANG Q H , TANG Z H , et al. Research on mask wearing detection method based on YOLOv5 enhancement model. Computer Engineering, 2022, 48 (4): 39- 49. doi: 10.19678/j.issn.1000-3428.0061502
13	邓天民, 谭思奇, 蒲龙忠. 基于改进YOLOv5s的交通信号灯识别方法. 计算机工程, 2022, 48 (9): 55- 62. doi: 10.19678/j.issn.1000-3428.0062843
	DENG T M , TAN S Q , PU L Z . Traffic light recognition method based on improved YOLOv5s. Computer Engineering, 2022, 48 (9): 55- 62. doi: 10.19678/j.issn.1000-3428.0062843
14	JIA Z H , SUN S K , LIU G C . Real-time traffic sign detection based on weighted attention and model refinement. Neural Processing Letters, 2023, 55 (6): 7511- 7527. doi: 10.1007/s11063-023-11271-8
15	LI S , WANG S , WANG P . A small object detection algorithm for traffic signs based on improved YOLOv7. Sensors (Basel, Switzerland), 2023, 23 (16): 7145. doi: 10.3390/s23167145
16	WANG C Y, BOCHKOVSKIY A, LIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D.C., USA: IEEE Press, 2023: 7464-7475.
17	REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. [2023-11-05]. http://arxiv.org/abs/1804.02767v1.
18	GE Z, LIU S T, WANG F, et al. YOLOX: exceeding YOLO series in 2021[EB/OL]. [2023-11-05]. http://arxiv.org/abs/2107.08430v2.
19	YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions[EB/OL]. [2023-11-05]. http://arxiv.org/abs/1511.07122v3.
20	WANG P Q, CHEN P F, YUAN Y, et al. Understanding convolution for semantic segmentation[C]//Proceedings of the IEEE Winter Conference on Applications of Computer Vision. Washington D.C., USA: IEEE Press, 2018: 1451-1460.
21	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Washington D.C., USA: IEEE Press, 2016: 770-778.
22	YANG G Y, LEI J, ZHU Z K, et al. AFPN: asymptotic feature pyramid network for object detection[EB/OL]. [2023-11-05]. http://arxiv.org/abs/2306.15988v2.
23	LIU S T, HUANG D, WANG Y H. Learning spatial fusion for single-shot object detection[EB/OL]. [2023-11-05]. http://arxiv.org/abs/1911.09516v2.
24	HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D.C., USA: IEEE Press, 2018: 7132-7141.
25	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[EB/OL]. [2023-11-05]. https://arxiv.org/abs/1706.03762.
26	ZHANG Y F , REN W Q , ZHANG Z , et al. Focal and efficient IOU loss for accurate bounding box regression. Neurocomputing, 2022, 506, 146- 157. doi: 10.1016/j.neucom.2022.07.042
27	ZHANG J, ZOU X, KUANG L D, et al. CCTSDB 2021: a more comprehensive traffic sign detection benchmark[EB/OL]. [2023-11-05]. https://centaur.reading.ac.uk/106129/1/12-23.pdf.
28	ZHANG J M , HUANG M T , JIN X K , et al. A real-time Chinese traffic sign detection algorithm based on modified YOLOv2. Algorithms, 2017, 10 (4): 127. doi: 10.3390/a10040127
29	LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[EB/OL]. [2023-11-05]. https://arxiv.org/abs/1512.02325.
30	BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4: optimal speed and accuracy of object detection[EB/OL]. [2023-11-05]. http://arxiv.org/abs/2004.10934v1.

[1]	HU Chaoju, GUO Fengyi. MODF Port State Detection Algorithm Based on Improved YOLOv7 [J]. Computer Engineering, 2025, 51(2): 78-85.
[2]	ZHANG Huan, WANG Chen, SHAN Jingdong, QIU Runhe. Elevator Safety Risk Prediction Based on Domain Adaptation and Attention Mechanism [J]. Computer Engineering, 2025, 51(2): 86-93.
[3]	ZHANG Yuan, LÜ Defang, MENG Jianjun, QI Wenzhe. Defect Detection of Rail Fasteners Based on Double Attention and GSSN Lightweight [J]. Computer Engineering, 2025, 51(2): 289-299.
[4]	ZHANG Xingpeng, HE Dong, YANG Mo, YE Hangbin. Nucleus Segmentation Based on Multiscale Attention and Data Augmentation [J]. Computer Engineering, 2025, 51(2): 387-396.
[5]	LUO Xudong, YUAN Di, CHANG Xiaojun, HE Zhenyu. Underwater Target Tracking Based on Uncertainty-Inspired Image Enhancement [J]. Computer Engineering, 2025, 51(1): 11-19.
[6]	ZHOU Xueyang, FU Qiming, CHEN Jianping, CHEN Yanming, LU You, WANG Yunzhe. Document-Level Relation Extraction Method Based on Evidence and Graph Inference: A Case Study of Medical Relationships [J]. Computer Engineering, 2025, 51(1): 106-117.
[7]	XIAO Chaoen, LI Zifan, ZHANG Lei, WANG Jianxin, QIAN Siyuan. Differential Cryptanalysis Based on Transformer Model and Attention Mechanism [J]. Computer Engineering, 2025, 51(1): 156-163.
[8]	YANG Lisha, LI Maojun, HU Jianwen, WANG Dingxiang. Strip Steel Surface Defect Detection Algorithm Based on Improved YOLOv7-tiny [J]. Computer Engineering, 2025, 51(1): 208-215.
[9]	HU Yongtao, HUANG Hongqiong. Multi-Branch Clothes-Changing Person Re-Identification with Feature Fusion and Channel Attention [J]. Computer Engineering, 2025, 51(1): 225-234.
[10]	HUO Jiuyuan, SU Hongrui, WU Zeyu, WANG Tingjuan. Road Traffic Small Target Vehicle Detection Algorithm Based on Improved YOLOv8 [J]. Computer Engineering, 2025, 51(1): 246-257.
[11]	ZHENG Yazhou, LIU Wanping, HUANG Dong. A BERT-CNN-GRU Detection Method Based on Attention Mechanism [J]. Computer Engineering, 2025, 51(1): 258-268.
[12]	WANG Qian, ZHANG Junhua, WANG Zetong, LI Bo. X2S-Net: Three-Dimensional Reconstruction of Spine Based on Biplanar X-Rays [J]. Computer Engineering, 2025, 51(1): 277-286.
[13]	LI Mengkun, YUAN Chen, WANG Qi, ZHAO Chong, CHEN Jingxuan, LIU Lifeng. Research on Online Listening Behavior Recognition Model Based on Improved YOLOv8 Algorithm [J]. Computer Engineering, 2025, 51(1): 287-294.
[14]	LIU Zhong, TANG Hong, WANG Ningzhe, ZHU Chuanrun. Text Summarization Method Incorporating RNN and Sparse Self-Attention [J]. Computer Engineering, 2025, 51(1): 312-320.
[15]	LIU Zhaowei, FANG Yanhong, ZHENG Mingyu, SUO Bin. Lung Disease Diagnosis Method Based on Attention Mechanism and Multi-tasking [J]. Computer Engineering, 2025, 51(1): 332-342.

Please choose a citation manager

Content to export