Small traffic sign detection method based on improved YOLOv8m

Xu Yang; Yixin Su; Bingrong Xu; Jicheng Yu

doi:10.1117/12.3032837

5 July 2024 Small traffic sign detection method based on improved YOLOv8m

Xu Yang, Yixin Su, Bingrong Xu, Jicheng Yu

Proceedings Volume 13184, Third International Conference on Electronic Information Engineering and Data Processing (EIEDP 2024); 131846M (2024) https://doi.org/10.1117/12.3032837
Event: 3rd International Conference on Electronic Information Engineering and Data Processing (EIEDP 2024), 2024, Kuala Lumpur, Malaysia

Abstract

The recognition of traffic signs plays a crucial role in maintaining road safety and facilitating the smooth flow of traffic. This paper introduces a novel small traffic sign recognition algorithm based YOLOv8m network structure. Initially, a new feature fusion structure, SSF, is proposed to replace the Neck part of YOLOv8, incorporating a lower-scale detection head with minimal cost and better leveraging the correlation of pyramid layers for enhanced multi-scale feature fusion. Following this, the C2f module is augmented with a self-attention operation, AttnConv, to boost the network's proficiency in extracting features from small traffic signs. Finally, the use of MPDIoU effectively reduces computation and accelerates the network's rapid fitting. Experimental validation on the processed TT100K dataset has shown that the improved model achieved a mean Average Precision(mAP) of 90.6%, which is a 2.4% improvement over the original YOLOv8m model and surpasses the current state-of-the-art, effectively proving the superiority of the proposed model in the recognition of small traffic signs.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Xu Yang, Yixin Su, Bingrong Xu, and Jicheng Yu "Small traffic sign detection method based on improved YOLOv8m", Proc. SPIE 13184, Third International Conference on Electronic Information Engineering and Data Processing (EIEDP 2024), 131846M (5 July 2024); https://doi.org/10.1117/12.3032837

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Object detection

Feature fusion

Feature extraction

Detection and tracking algorithms

Data modeling

Education and training

Head

Show All Keywords

Keywords/Phrases

Search In:

Publication Years