Sage Journals: Discover world-class research

Abstract

License plate location is one of the main research topics in intelligent traffic management. The whole process of license plate location and recognition is as follows: preprocessing the image and get the location of the license plate roughly by combining morphology; screening the license plates that match the features; obtaining the location of the license plate accurately, and finally completing the character recognition. The traditional license plate location methods based on edge, color, texture and machine learning need to extract complex features to license plate image, which can not only lead to overfitting or dimensionality in the training process, but also be affected by illumination, road environment and image quality etc. This paper proposed an algorithm of vehicle license plate location based on convolutional neural network, which avoids the problem that the traditional location algorithms have to preprocess images and need rich experience to extract the feature of the samples. The experimental results show that the convolutional neural network has better performance in license plate location compared with BP neural network and support vector machine (SVM), and also prove that deep learning has wide application in the field of intelligent transportation.

Keywords

License plate location feature map convolutional neural network recognition rate

Get full access to this article

View all access options for this article.

References

Bengio

Courville

and Vincent

, Representation learning: a review and new perspectives, IEEE Transactions on Pattern Analysis and Machine Intelligence 35(8) (2013), 1798–1828.

Silver

Huang

Maddison

C.J.

et al., Mastering the game of go with deep neural networks and tree search, Nature 529(7587) (2016), 484–489.

Otsu

, A threshold selection method from gray-level histogram, IEEE Transactions on Systems, Man and Cybernetics 9(1) (1979), 652–655.

Zhang

Ren

et al., Spatial pyramid pooling in deep convolutional networks for visual recognition, IEEE Transactions on Pattern Analysis and Machine Intelligence 37(9) (2014), 1904–1916.

Krizhevsky

Sutskever

and Hinton

G.E.

, Image net classification with deep convolutional neural networks, in: International Conference on Neural Information Processing Systems, 2012, pp. 1097–1105.

Zeiler

M.D.

and Fergus

, Visualizing and understanding convolutional networks, in: European Conference on Computer Vision, Springer, Cham, 2014, pp. 818–833.

Zeiler

M.D.

and Fergus

, Stochastic pooling for regularization of deep convolutional neural networks, Xiv pre-print Xiv, 2013, 130–135.

Srivastava

Hinton

Krizhevsky

et al., Dropout: a simple way to prevent neural networks from overfitting, Journal of Machine Learning Research 15(1) (2014), 1929–1958.

Girshick

Donahue

Darrell

et al., Rich feature hierarchies for accurate object detection and semantic segmentation, in: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2014, pp. 580–587.

10.

Simonyan

and Zisserman

, Very deep convolutional networks for largescale image recognition, Xiv pre-print Xiv, 2014, 1409–1556.

11.

Zhang

Ren

et al., Deep residual learning for image recognition, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 770–778.

12.

Glorot

and Bengio

, Understanding the difficulty of training deep feed forward neural networks, Journal of Machine Learning Research 9(1) (2010), 249–256.

13.

Srivastavar

Greff

Schmid

H.J.

et al., Training very deep networks, Advances in Neural Information Processing Systems (2015), 2377–2385.

14.

Khalajzadeh

Mansouri

Teshnehlab

et al., Hierarchical structure based convolutional neural network for face recognition, International Journal of Computational Intelligence and Applications 12(3) (2013), 1–15.

A new algorithm of vehicle license plate location based on convolutional neural network

Abstract

Keywords

Get full access to this article

References