摘要： 在疲劳驾驶检测中,对眼部状态的判断是关键的步骤之一。为了对眼部状态进行有效的识别,提出了一种新颖的眼部状态识别 方法。该方法用眼部图像中的某些点的纹理单元的NTU值作为输入特征值,用径向基函数神经网络(RBF)作为分类器。为了进一步提高分类的准确性,又采用了Bagging方法。试验结果表明,该方法易于实现,准确度高,速度快,不受光照条件的影响,可以应用于实际。

关键词: 眼部状态识别, 纹理单元, RBF神经网络, Bagging算法

Abstract: This paper presents a new method of eye state recognition. Firstly, it uses NTU as the input eigenvalue, which is picked up from texture character of eye images. RBF neural network is used as classifer. In order to improve the precision of the RBF neural network models, Bagging algorithm is used to build an integration neural network model for eye state recognition. Some experiments to make sure that the method works effective are performed.

中图分类号: 

• TP391.41