Verification of the Effectiveness of the Online Tuning System for Unknown Person in the Awaking Behavior Detection System
Verification of the Effectiveness of the Online Tuning System for Unknown Person in the Awaking Behavior Detection System
复制标题
清醒行为检测系统中未知人员在线调优系统的有效性验证
DOI：
10.1007/978-3-642-02481-8_39
复制
发表时间：
2009
期刊：
影响因子：
--
通讯作者：
F. Takeda
中科院分区：
文献类型：
--
作者：
H. Satoh;F. Takeda
我们开发了一种使用神经网络（缩写为 NN）的清醒行为检测系统。然而，与有学问的人相比，未知人的检测能力还不够。在本研究中，为了提高对未知人员的检测能力，我们应用了在线调整系统，该系统利用神经网络的持续学习来检测系统。在在线调优系统中，仅使用新目标人的少量附加数据进行连续学习，其中初始学习中收敛的神经网络的权重用作连续学习的初始权重。在本文中，为了验证在线调优系统的能力，我们比较了收敛初始学习和收敛在线调优的检测能力。
We have developed an awaking behavior detection system using a neural network (abbreviated as NN). However, the detection ability of unknown people is not sufficient with compared to that of learned people. In this research, to improve the detection ability of unknown people, we apply an online tuning system using a continuous learning of the NN for the detection system. In the online tuning system, only a few additional data of a new objective person are used for the continuous learning, where the weights of the NN converged in the initial learning are used as the initial weights for the continuous learning. In this paper, to verify an ability of the online tuning system, we compare detection ability of the converged initial learning with that of the converged online tuning.