Fusion of Spatial Information for YOLOv7 Traffic Sign Detection

doi:10.3969/j.issn.1006-2475.2025.10.002

Abstract

Abstract: Abstract: During the detection process of traffic signs， due to the influence of weather and light intensity， problems such as false detections and missed detections occur during detection. To solve this problem， a traffic sign detection algorithm combining spatial information is proposed. Firstly， coordinate convolution is used on network to enhance sensitivity of the network to coordinate position information. Additionally， the incorporation of a coordinate attention mechanism into the backbone features enables better focus on spatial location information at fusion points. Moreover， the feature fusion process utilizes a multi-scale weighted network and pyramid pooling， leveraging weighted calculations and skip connections to enhance semantic information fusion between low-level and high-level layers. Lastly， the adoption of the SIoU loss function enhances target positioning accuracy. The experimental results on the CCTSDB2021 and GTSDB datasets demonstrate that this method achieved mean Average Precision (mAP) values of 84.9% and 98.5% respectively. Compared with mainstream detection models， it shows significant improvement—exceeding the original model by 5.39 percentage points and 1.67 percentage points—thus enhancing the detection accuracy of traffic signs.

Key words: Key words: traffic sign detection, coordinate convolution, attention mechanism, multi-scale fusion, SIoU loss function

CLC Number:

中图分类号：TP391.41

SHI Hongyu, ZHANG Zheyu, DU Wen, LI Yi. Fusion of Spatial Information for YOLOv7 Traffic Sign Detection[J]. Computer and Modernization, 2025, 0(10): 7-13.

References

［1］《中国公路学报》编辑部. 中国汽车工程学术研究综述·2023［J］. 中国公路学报， 2023，36（11）:1-192.
［2］陈飞，刘云鹏，李思远. 复杂环境下的交通标志检测与识别方法综述［J］. 计算机工程与应用， 2021，57（16）:65-73.
［3］廖聪，郭凰，赵茂军，等. 基于图像增强和SKNet的交通标志识别［J］. 计算机与现代化， 2023（3）:23-28.
［4］ GIRSHICK R， DONAHUE J， DARRELL T， et al. Rich feature hierarchies for accurate object detection and semantic segmentation［C］// 2014 IEEE Conference on Computer Vision and Pattern Recognition. IEEE， 2013:580-587.
［5］ REN S Q， HE K M， GIRSHICK R， et al. Faster R-CNN: Towards real-time object detection with region proposal networks［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2015，39（6）:1137-1149.
［6］ HE K M， GEORGIA G， PIOTRL D， et al. Mask R-CNN［C］// IEEE International Conference on Computer Vision. IEEE， 2017:386-397.
［7］项新建，姚佳娜，黄炳强，等. 复杂光照下的交通标志检测与识别［J］.计算机辅助设计与图形学学报， 2023，35（2）:293-302.
［8］ LIU W， ANGUELOV D， ERHAN D， et al. SSD: Single shot multiBox detector［C］// Computer Vision-Computer Research Repository 2016. Springer， 2016:21-37.
［9］ LECUN Y， BENGIO Y， HINTON G. Deep learning［J］. Nature， 2015，521（7553）:436-444.
［10］郭璠，张泳祥，唐琎，等. YOLOv3-A：基于注意力机制的交通标志检测网络［J］. 通信学报， 2021，42（1）:87-99.
［11］ QU S M， YANG X Y， ZHOU H F， et al. Improved YOLOv5-based for small traffic sign detection under complex weather［J］. Scientific Reports， 2023，13（1）. DOI: 10.
1038/s41598-023-42753-3.
［12］ WANG X， TIAN Y， ZHENG K， et al. C2Net-YOLOv5: A bidirectional Res2Net-based traffic sign detection algorithm［J］. Computers， Materials & Continua， 2023，77（2）:1949-1965.
［13］ LUO J P， WANG Z F. A low latency traffic sign detection model with an automatic data labeling pipeline［J］. Neural Computing and Applications， 2022，34:15499-15512.
［14］ HUANG M T， WAN Y Y， GAO Z W， et al. Real-time traffic sign detection model based on multi-branch convolutional reparameterization［J］. Journal of Real Time Image Process， 2023，20（3）:1-16.
［15］ 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 （CVPR）. IEEE， 2022:7464-7475.
［16］ LIU R， LEHMAN J， MOLINO P， et al. An intriguing failing of convolutional neural networks and the CoordConv solution［C］// Proceedings of the 32nd International Conference on Neural Information Processing Systems（NeurIPS）. ACM， 2018:9628-9639.
［17］ LI C Y， LI L L， GENG Y F， et al. YOLOv6 v3.0: A full-scale reloading［J］ arXiv preprint arXiv:2023.05586v1，2020.
［18］ TAN M X， PANG R M， LE Q V. EfficientDet: Scalable and efficient object detection［C］// Conference on Computer Vision and Pattern Recognition（CVPR）. IEEE， 2020:
10778-10787.
［19］ LIN T Y， DOLLÁR P， GIRSHICK R， et al. Feature pyramid networks for object detection［C］// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE， 2017:2117-2125.
［20］ LIU S， QI L， QIN H F， et al. Path aggregation network for instance segmentation［C］// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE， 2018:8759-8768.
［21］ HU J， LI S， SUN G. Squeeze-and-excitation networks［C］// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE， 2018:7132-7141.
［22］ WOO S， PARK J， LEE J Y， et al. CBAM: Convolutional block attention module［C］// Proceedings of the European Conference on Computer Vision（ECCV）. Springer， 2018:3-19.
［23］ HOU Q B， ZHOU D Q， FENG J S. Coordinate attention for efficient mobile network design［C］// Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE， 2021:13708-1371.
［24］ ZHENG Z H， WANG P， REN D W， et al. Enhancing geometric factors in model learning and inference for object detection and instance segmentation［J］. IEEE Transactions on Cybernetics， 2021，52（8）:8574-8586.
［25］ GEVORGYAN Z. SIoU loss: More powerful learning for bounding box regression［J］. arXiv preprint arXiv:2205.127
40， 2022.
［26］ ZHANG J M， ZOU X， KUANG L-D， et al. CCTSDB 2021: A more comprehensive traffic sign detection benchmark［J］. Human-centric Computing and Information Sciences， 2022，12. DOI: 10.22967/HCIS.2022.12.023.
［27］ HOUBEN S STALLKAMP J， SALMEN J， et al. Detection of traffic signs in real-world images: The German traffic sign detection benchmark［C］// International Symposium on Neural Networks. IEEE， 2013:1-8.

[1]	ZHOU Bangyuan1, XIN Guojiang1, LIANG Hao2, DING Changsong1. Pancreas Segmentation Based on Two-stage Network of Multiple Attention Mechanisms [J]. Computer and Modernization, 2025, 0(10): 67-72.
[2]	LIU Rongcheng1, XIN Guojiang1, ZHANG Yang1, ZHU Lei2. Natural Environment Tongue Image Segmentation Method Based on Improved Labv3+ Model [J]. Computer and Modernization, 2025, 0(10): 32-36.
[3]	GUO Jinhao, WANG Fengping, WANG Haoqi. PB-YOLOv7 Pedestrian Detection Method for Dense Scenes [J]. Computer and Modernization, 2025, 0(09): 14-19.
[4]	LIN Ruizi, YAO Da, DAI Xin, SHEN Guoyu, WANG Jiahui, WAN Weiguo. Facial Sketch Image Conversion Based on CycleGAN and Attention Mechanism [J]. Computer and Modernization, 2025, 0(09): 61-66.
[5]	LIU Cheng, FENG Guang. CNN-BiLSTM and LightGBM Stock Prediction Based on Dual Attention Mechanism [J]. Computer and Modernization, 2025, 0(09): 97-103.
[6]	JING Qingwu, CHEN Hongjun, GAO Di, ZHOU Meimei. Goal Driven Recommendation-oriented Dialog Generation Method [J]. Computer and Modernization, 2025, 0(08): 16-23.
[7]	SU Yansen, MOU Li. Improved Classroom Behavior Detection Algorithm for YOLOv8 [J]. Computer and Modernization, 2025, 0(08): 76-81.
[8]	GAO Lisha1, PENG Guozheng2, WANG Chong1, HAN Shuo1, XIANG Nan1, ZHANG Qizhe2, ZHU Yuefei3, CHENG Xu3. Dual-backbone Network Insulator Defect Detection Algorithm Based on YOLOv5 [J]. Computer and Modernization, 2025, 0(08): 82-88.
[9]	CAI Bohan, LIU Jun. Long-and Short-Term Air Pollutant Concentration Forecasting Based on Optimized Transformer [J]. Computer and Modernization, 2025, 0(07): 55-62.
[10]	JIAO Leyan, ZHU Xinjuan. Speech Cloning Method Based on Self-attention Mechanism Speaker Encoder And SA-Decoder [J]. Computer and Modernization, 2025, 0(07): 69-76.
[11]	RAO Hongyu1, CHEN Xin2, CHEN Sheng3, XIA Tian3, SHEN Li4, LUY Guangqiang5. Power Load Forecasting of Bi-LSTM Based on Improving Whale Algorithm [J]. Computer and Modernization, 2025, 0(07): 77-82.
[12]	WEI Qingsong1, WANG Xiaojun2, WEI Yuan1, ZENG Shangqi1, FANG Yuan1. HGAT: Multivariate Time Series Anomaly Detection Based on Hybrid Graph Attention Network [J]. Computer and Modernization, 2025, 0(07): 97-105.
[13]	XIE Peidong, CHENG Yuanzhi, XU Haotian. Hierarchical Classification Algorithm for Marine Organisms [J]. Computer and Modernization, 2025, 0(06): 21-27.
[14]	HAO Rongrong1, MA Qiaomei1, TAN Yajun1, SHI Huanyin2. Complex Fire Detection Based on Deformable Convolutions and Attention Mechanisms [J]. Computer and Modernization, 2025, 0(06): 79-85.
[15]	QIN Hongzheng1, 2, WANG Tonghan1, 2, JIA Huizhen1, 2. Image Quality Assessment with Multi-scale Interaction and Cross-channel Attention [J]. Computer and Modernization, 2025, 0(06): 86-91.