Research on application of meta-learning in traffic light recognition

Xiaole Su; Guiping Wang; Ruonan Yang; Siyu He; Jinzhuo Hu

doi:10.1117/12.2685686

28 July 2023 Research on application of meta-learning in traffic light recognition

Xiaole Su, Guiping Wang, Ruonan Yang, Siyu He, Jinzhuo Hu

Proceedings Volume 12716, Third International Conference on Digital Signal and Computer Communications (DSCC 2023); 127161F (2023) https://doi.org/10.1117/12.2685686
Event: Third International Conference on Digital Signal and Computer Communications (DSCC 2023), 2023, Xi'an, China

Abstract

Traffic light recognition is essential to intelligent vehicle information perception, and its accuracy directly affects traffic safety. Based on sorting out and analyzing the existing related research on traditional machine learning and deep learning in traffic light recognition, the basic principles of two standard algorithms in meta-learning are introduced in detail. Compared with traditional machine learning, using the MAML algorithm and Reptile algorithms in meta-learning to recognize traffic lights by selecting the WPI dataset and programming in PYTHON. The simulation results show that the recognition accuracy based on the meta-learning algorithm is higher than that based on the traditional machine learning algorithm; In the meta-learning algorithm, the Reptile algorithm outperforms the MAML algorithm.

Citation Download Citation

Xiaole Su, Guiping Wang, Ruonan Yang, Siyu He, and Jinzhuo Hu "Research on application of meta-learning in traffic light recognition", Proc. SPIE 12716, Third International Conference on Digital Signal and Computer Communications (DSCC 2023), 127161F (28 July 2023); https://doi.org/10.1117/12.2685686

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