Improved LeNet-5 Convolutional Neural Network Traffic Sign Recognition

Lifen Li ； Yangrui Wang

《International Core Journal of Engineering》 7卷4期 (2021/04) Pp. 114-121

https://doi.org/10.6919/ICJE.202104_7(4).0017

摘要

In order to improve the speed and accuracy of road traffic sign detection and recognition, the classical convolutional neural network architecture Lenet-5 is proposed to be improved and optimized. LeNet-5 network structure was improved, reducing unnecessary link and calculation to improve the real-time performance of algorithm, and combined with the feature of multi-layer fusion method, will be in the two layers of convolution layer extracts the features of the network are sent to all the connection layer, strengthen the characteristic power of expression, on the basis of meet the real time improve the identification accuracy of the algorithm and introduce Dropout mechanism at the same time, to avoid over fitting, improve the network generalization ability of the model. Experimental results show that the improved Lenet-5 convolutional neural network proposed in this paper achieves the desired effect in terms of accuracy and real-time performance.

關鍵字

LeNet-5 ； Multilayer Feature Fusion ； Convolutional Neural Network ； Traffic Sign Recognition

參考文獻

Dalal N, Triggs B. Histograms of oriented gradients for human detection[C]//2005 IEEE computer society conference on computer vision and pattern recognition (CVPR'05). IEEE, 2005, 1: 886-893.

Schügerl P, Sorschag R, Bailer W, et al. Object re-detection using SIFT and MPEG-7 color descriptors [C]// International Workshop on Multimedia Content Analysis and Mining. Springer, Berlin, Heidelberg, 2007: 305-314..

ZHU Y, WANG X, YAO C, et al. Traffic sign classification using two-layer image representation[C]// 2013 IEEE International Conference on Image Processing. Melbourne: IEEE, 2013: 3755-3759.

Lecun Y, Bottou L, Bengio Y, et al.Gradient-Based Learning Applied to Document Recognition[J]. Proceedings of the IEEE, 1998,86(11):2278-2324.

SERMANET P, LECUN Y.Traffic sign recognition with multi-scale convolutional networks[C]//The 2011International Joint Conference on Neural Networks (IJCNN) .Washington DC:IEEE Computer Society, 2011:2809-2813.