论文部分内容阅读
在现有的数据条件下,疲劳判定的本质问题是分类问题,BP神经网络是一种经常应用于人脸识别领域的分类算法,其不需要一套人为确定的规则,并可以根据有代表性的样本自我学习。论文采用BP神经网络算法进行飞行员的疲劳识别,首先定位出飞行员面部和头部特征点,然后根据特征点计算出相应的特征向量,把这些特征向量输入BP神经网络分类器,最后利用神经网络模糊分类的优点,进行飞行员疲劳判定。
Under the existing data conditions, the essential problem of fatigue assessment is the classification problem. BP neural network is a classification algorithm often applied in the field of face recognition. It does not need a set of artificially determined rules and can be based on the representative Self-learning samples. In this paper, the BP neural network algorithm is used to detect the pilot’s fatigue. First, the pilot’s face and head feature points are located. Then, the corresponding feature vectors are calculated according to the feature points. These feature vectors are input to the BP neural network classifier. Finally, The advantages of classification, pilot fatigue assessment.