| 树状卷积神经网络的车标识别应用 |
| |
| 引用本文： | 吴章辉,李志清,杨晓玲,刘雨桐. 树状卷积神经网络的车标识别应用[J]. 计算机系统应用, 2017, 26(10): 166-171 |
| |
| 作者姓名： | 吴章辉 李志清 杨晓玲 刘雨桐 |
| |
| 作者单位： | 湘潭大学 信息工程学院学院, 湘潭 411105,湘潭大学 信息工程学院学院, 湘潭 411105,湘潭大学 信息工程学院学院, 湘潭 411105,湘潭大学 信息工程学院学院, 湘潭 411105 |
| |
| 摘 要： | 为了提高在自然环境下车标识别率，提出一种多通路树状结构的卷积神经网络模型.该模型采用多通路树状结构，在传统卷积网络单一种类卷积核的卷积层上，使用多种类型的卷积核进行卷积操作，并且采用树状网络结构.通过对每个通路的顶层提取特征，作为全连接层的输入，进行车标的分类任务.通过理论分析和实验表明，与传统的卷积神经网络训练获得的分类器相比，车标识别率提升至98.43%.
|
| 关 键 词： | 深度学习 卷积神经网络 特征图 车标识别 |
| 收稿时间： | 2017-01-12 |
Vehicle Logo Recognition Using Tree-Based Convolution Neural Network |
| |
| WU Zhang-Hui,LI Zhi-Qing,YANG Xiao-Ling and LIU Yu-Tong. Vehicle Logo Recognition Using Tree-Based Convolution Neural Network[J]. Computer Systems& Applications, 2017, 26(10): 166-171 |
| |
| Authors: | WU Zhang-Hui LI Zhi-Qing YANG Xiao-Ling LIU Yu-Tong |
| |
| Affiliation: | The College of Information Engineering, Xiangtan University, Xiangtan 411105, China,The College of Information Engineering, Xiangtan University, Xiangtan 411105, China,The College of Information Engineering, Xiangtan University, Xiangtan 411105, China and The College of Information Engineering, Xiangtan University, Xiangtan 411105, China |
| |
| Abstract: | In order to improve the recognition rate of vehicle in natural situations, this paper proposes a vehicle logo recognition modal based on a multi-path tree structure convolutional neural networks, which modal with different convolution kernel in the same convolutions, namely T-CNN. Firstly, different layer convolution features are obtained and are joined together as the input of the fully connected layer to get classifiers. Compared with the traditional method, the theoretical analysis and simulation results show that T-CNN can increase the recognition accuracy up to 98.43%. |
| |
| Keywords: | deep leaning convolutional neural network(CNN) feature map vehicle logo recognition |
|
| 点击此处可从《计算机系统应用》浏览原始摘要信息 |
|
点击此处可从《计算机系统应用》下载全文 |
|
