NSSC OpenIR  > 空间技术部
Optimization of Convolutional Neural Network Target Recognition Algorithm
Guo, Chen; Jiang, Yuanyuan
作者部门空间技术部
会议录名称14TH INTERNATIONAL CONFERENCE ON WIRELESS COMMUNICATIONS, NETWORKING AND MOBILE COMPUTING (WICOM 2018)
2018
卷号306
页码426-433
语种英语
ISSN号2475-8841
ISBN号978-1-6059-5578-0
摘要This paper proposes an optimized convolutional neural network target recognition algorithm for the problem of low recognition rate of synthetic aperture radar (SAR) target training, under the condition of insufficient tag data, translation, rotation and complexity. In order to overcome the shortage of tag data, the convolutional neural network is initialized with a feature set, obtained by principal component analysis (PCA) unsupervised training. In order to improve the training speed while avoiding overfitting, Rectified Linear Unit (ReLU) function is used as the activation function. In order to enhance robustness and reduce the effect of down sampling on feature representation, this work uses a maximum probability sampling method and normalizes the local contrast of feature after convolution layers. The experimental result shows that, compared with traditional convolutional neural network, this approach achieves a higher recognition rate for SAR target and better robustness to various image deformation and complex background.
关键词Convolutional neural networks Deep learning principal component analysis
会议名称14th International Conference on Wireless Communications, Networking and Mobile Computing (WiCOM)
会议日期SEP 18-20, 2018
会议地点Kunming, PEOPLES R CHINA
收录类别CPCI
文献类型会议论文
条目标识符http://ir.nssc.ac.cn/handle/122/6680
专题空间技术部
推荐引用方式
GB/T 7714
Guo, Chen,Jiang, Yuanyuan. Optimization of Convolutional Neural Network Target Recognition Algorithm[C],2018:426-433.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
201826290-41358-1-SM(1147KB)会议论文 开放获取CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[Guo, Chen]的文章
[Jiang, Yuanyuan]的文章
百度学术
百度学术中相似的文章
[Guo, Chen]的文章
[Jiang, Yuanyuan]的文章
必应学术
必应学术中相似的文章
[Guo, Chen]的文章
[Jiang, Yuanyuan]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。