Abstract
I approach the traffic signs classification problem with a convolutional neural network implemented in TensorFlow reaching 99.33% accuracy. The highlights of this solution would be data pre-processing, data augmentation pipeline, pre-training and skipping connections in the network. I am using Python as programming language and TensorFlow as a fairly low-level machine learning framework.
Similar content being viewed by others
References
P. Sermanet and Y. LeCun, “Traffic sign recognition with multi-scale convolutional networks,” in Proc. Int. Joint Conf. on Neural Networks (IJCNN’11) (San Jose, 2011).
M. Haloi, “Traffic sign classification using deep inception based convolutional networks” (2015). arXiv:1511.02992, 2015.
D. C. Ciresan, U. Meier, J. Masci, and J. Schmidhuber, “A committee of neural networks for traffic sign classification,” in Proc. Int. Joint Conf. on Neural Networks (IJCNN) (San Jose, 2011), pp. 1918–1921.
J. Stallkamp, M. Schlipsing, J. Salmen, and C. Igel, “The German traffic sign recognition benchmark: a multi-class classification competition,” in Proc. Int. Joint Conf. on Neural Networks (San Jose, 2011).
Author information
Authors and Affiliations
Corresponding author
Additional information
The article is published in the original.
Aliaksei (Alex) Staravoitau was born in Minsk, Belarus, in 1988. He received the B.E. degree in Mathematics from the Belarusian State University, Minsk, Belarus, in 2011, and is a first-year MSc student in computer science at the Belarusian State University of Informatics and Radioelectronics. Aliaksei has been an independent researcher since 2015, focusing mainly on applications of statistics and machine learning.
Rights and permissions
About this article
Cite this article
Staravoitau, A. Traffic Sign Classification with a Convolutional Network. Pattern Recognit. Image Anal. 28, 155–162 (2018). https://doi.org/10.1134/S1054661818010182
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1054661818010182