Fatigue driving detection method based on deep integrated optimization fuzzy classifier

The invention provides a deep integrated optimization fuzzy classifier-based fatigue driving detection method, which comprises the following steps of: inputting preprocessed data into a deep integrated optimization fuzzy classifier to obtain a first output vector and a first fatigue driving detection result, and randomly selecting a first fuzzy rule at the same time; judging whether the first fatigue driving detection result reaches the preset classification precision or not, then judging whether the second fatigue driving detection result reaches the preset classification precision or not, if yes, outputting the fatigue driving detection result, and if not, continuously increasing splicing of output vectors and fuzzy rule cumulative multiplication operation. According to the method, the problems of unbalanced fatigue driving data and poor model interpretability are solved. 本发明提供了一种基于深度集成优化模糊分类器的疲劳驾驶检测方法，该方法包括：将经预处理后的数据输入至深度集成优化模糊分类器，得到第一输出向量和第一疲劳驾驶检测结果，并同时随机选择第一模糊规则；先对第一疲劳驾驶检测结果是否达到预设的分类精度进行判断，然后再对第二疲劳检测结果是否