Open AccessResearch on Vehicle Recognition Algorithm based on Convolution Neural Network
Author(s) -
Pai Zhang,
Hanqing Chen,
Qinrui Li
Publication year - 2021
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 - 10.1088/1742-6596/1865/4/042117
Subject(s) - computer science , pooling , artificial intelligence , dropout (neural networks) , convolution (computer science) , convolutional neural network , basis (linear algebra) , artificial neural network , pattern recognition (psychology) , layer (electronics) , algorithm , machine learning , mathematics , chemistry , geometry , organic chemistry
In order to effectively identify vehicle types in intelligent transportation system, based on the analysis of Inception V3 model, a deep learning model of vehicle classification based on transfer learning theory is proposed in this paper. In this model, the last full connection layer is removed on the basis of Inception V3 model, and the parameter optimization layer is added, and then Dropout and global average pooling layer are adopted. Theoretical analysis and experimental results show that the performance of this model is better than that of VGG-16-based vehicle classification model, Xception-based vehicle classification model and Resnet-50-based vehicle classification model. The experimental results show that the training accuracy of the method proposed in this paper is more than 96.48% and the test accuracy is more than 83.86%.