Algorithm improvement for traffic sign detection based on YOLOv8

Dongmei Ma; Xuelong Lyu; Qirong Zhu

doi:10.1007/s11801-026-4136-y

Optoelectronics Letters ›› 2026, Vol. 22 ›› Issue (2) :92 -97. DOI: 10.1007/s11801-026-4136-y

Article

research-article

Algorithm improvement for traffic sign detection based on YOLOv8

Author information +

History +

PDF

Abstract

An improved algorithm for traffic sign detection based on YOLOv8 is proposed. Firstly, YOLOv8n is used as the base model of the network, the inverted residual mobile block and exponential moving average (iRMB_EMA) attention mechanism is used to improve the model’s ability to perceive small targets, which reduces the leakage detection phenomenon of the model, convolution (Conv) is upgraded to receptive-field attention convolution (RFAConv), which improves the model’s ability to deal with details and complexity in the image, the idea of adaptive spatial feature fusion (ASFF) is introduced in the detection head, and the small target detection layer, a four-head detection head is designed to improve the model’s ability to detect small targets, solves the case of feature loss due to cross-scale fusion, and use of the Inner-minimum points distance intersection over union (MPDIoU) loss function provides a more accurate loss metric by calculating the distance of key points between the predicted and true frames. The experimental results of this algorithm on the public dataset CCTSDB show that the improved model mean average precision (mAP) reaches 82.6%, which is 4% higher than the YOLOv8n. The experimental results of dataset TT100k show that the mAP reaches 84.5%, which is 7.1% higher than the YOLOv8n. This algorithm effectively improves the problem of detail perception and leakage of the model in the detection of small targets, and has a significant detection effect compared to other algorithms.

Keywords

Cite this article

Download citation ▾

Dongmei Ma, Xuelong Lyu, Qirong Zhu. Algorithm improvement for traffic sign detection based on YOLOv8. Optoelectronics Letters, 2026, 22(2): 92-97 DOI:10.1007/s11801-026-4136-y

登录浏览全文

4963

注册一个新账户忘记密码

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Mao W, Zhu J L, Wu Xet al. . A review of YOLO-based target detection research for autonomous driving. Computer engineering and application. 2022, 58(15): 68-77[J]

[2]	Yin J H, Qu S J, Yao Z Ket al. . Traffic sign recognition model in hazy weather based on YOLOv5. Computer applications. 2022, 42(09): 2876-2884[J]

[3]	GUO Z, SHI Y, AHMAD I. Design of smart citrus picking model based on mask RCNN and adaptive threshold segmentation[J]. PeerJ computer science, 2024: 10e1865.

[4]	Liu Z M, Wang J R, Li Jet al. . A novel multiple targets detection method for service robots in the indoor complex scenes. Intelligent service robotics. 2023, 16(4): 453-469. J]

[5]	Suliman A, Shabana H, Muhammad Iet al. . An efficient pest detection framework with a medium-scale benchmark to increase the agricultural productivity. Sensors. 2022, 22(24): 9749. J]

[6]	Ke X. Application of portrait recognition system for emergency evacuation in mass emergencies. Journal of intelligent systems. 2021, 301893-902. J]

[7]	JAVED S, RAMESH B T, KHUSH A, et al. Hexacopter-based modern remote sensing using the YOLO algorithm[J]. Advances in science and technology, 2023: 668075–668084.

[8]	Wang C Y, Bochkovskiy A, Liao H. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real time object detectors. IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 18–22, 2023, Vancouver, Canada. 2023, New York, IEEE74647475[C]

[9]	Xiao M, Gong Y F, Wang H Det al. . Defect detection of light guide plate based on improved YOLOv5 networks. Optoelectronics letters. 2024, 20(09): 560-567. J]

[10]	Yuan H C, Tao L. Detection and recognition of fish in electronic monitoring data of commercial fishing vessels based on improved YOLOv8. Journal of Dalian Ocean University. 2023, 38(03): 533-542[J]

[11]	Xiong E J, Zhang R F, Liu Y Het al. . Ghost-YOLOv8 detection algorithm for traffic signs. Computer engineering and applications. 2023, 59(20): 200-207[J]

[12]	Song P, Zhao L, Li Het al. . RSE-YOLOv8: an algorithm for underwater biological target detection. Sensors (Basel, Switzerland). 2024, 24(18): 6030. J]

[13]	ZHU M D, LIU W T, XIAO J B, et al. Defect detection in steel using a hybrid attention network[J]. Sensors (Basel, Switzerland), 2023, 23(15).

[14]	Gao Y, Li Z, Wang Yet al. . A novel YOLOv5es based on lightweight small object detection head for PCB surface defect detection. Scientific reports. 2024, 14123650. J]

[15]	Ma S L, Xu Y. Bounding box regression loss function for minimum point distance and its applications. Small microcomputer systems. 2024, 45112695-2701[J]

[16]	HU Z C, WANG Y, WANG J P, et al. Improved lightweight rebar detection network based on YOLOv8s algorithm[J]. Advances in computer, signals and systems, 2023, 7(10).

[17]	ZHANG J, 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.