Open AccessLicense Plate Location and Recognition on Neural Network
Author(s) -
Zhilong He,
Zhongjun Xiao,
Zhiguo Yan
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1642/1/012012
Subject(s) - convolutional neural network , license , computer science , preprocessor , artificial neural network , artificial intelligence , pattern recognition (psychology) , convolution (computer science) , computer vision , time delay neural network , character recognition , neocognitron , layer (electronics) , mobile device , image (mathematics) , chemistry , organic chemistry , operating system
This paper takes autonomous driving technology as the research background, and aims to improve the efficiency of the traffic system in real life, and proposes a neural network recognition method for license plates. The method realizes accurate location and recognition of LP (license plate) by constructing two neural recognition networks. For the LP location, a 11-layer convolution recognition network (CNN) is constructed. The basic step is to roughly locate the LP after preprocessing the image, and use the neural network to accurately locate the LP. For LP recognition, the 10-layer convolutional neural network constructed by inputting the segmented LP is used to realize character recognition, and the output result is obtained. The method is efficient and accurate, and can be applied to handheld devices in transportation systems.