Gradient and SVM based Biometric Identification using Human Body Communication | |
Meng Xia; Jingjing Ma; Jingzhen Li; Yuhang Liu; Yicheng Zeng; Zedong Nie | |
2016 | |
会议名称 | 2016 IEEE International Conference of Online Analysis and Computing Science (ICOACS 2016) |
会议地点 | 中国重庆 |
英文摘要 | To investigate biometric identification based on human body communication at 300KHz-1.5GHz. A feasibility study was done with 10 volunteers including 6 men and 4 women. The age range of the volunteers is 23-27 years, weigh 45-75 kilogram, and stand 153-180 centimeter in, all volunteers are healthy. The measurement was done 9 times in 3 days, and a total 2,880,000 measurement data has been obtained. Matrix transform was employed to extract gradient from measurement data. The gradients as an individual trait were analyzed by support vector machines (SVM) including C-SVM and nu-SVM with linear function, polynomial, and radial basis function (RBF) as kernel function. Our experimental results show that, when the C-SVM with RBF as kernel function is used, the correct identification rate (CIR) of 99.5% is achieved, the area under the curve (AUC) of receiver operating characteristic (ROC) reaches 0.9999 and the equal error rate (EER) is 0.11%. |
收录类别 | EI |
语种 | 英语 |
内容类型 | 会议论文 |
源URL | [http://ir.siat.ac.cn:8080/handle/172644/10554] |
专题 | 深圳先进技术研究院_医工所 |
作者单位 | 2016 |
推荐引用方式 GB/T 7714 | Meng Xia,Jingjing Ma,Jingzhen Li,et al. Gradient and SVM based Biometric Identification using Human Body Communication[C]. 见:2016 IEEE International Conference of Online Analysis and Computing Science (ICOACS 2016). 中国重庆. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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。
修改评论