基于语谱图的管制员疲劳状态检测研究  

Research on controller fatigue state detection based on speech spectrogram

作　　者：杨昌其[1] 冯筱晴 张雨萱 蔡子牛 YANG Changqi;FENG Xiaoqing;ZHANG Yuxuan;CAI Ziniu(College of Air Traffic Management,Civil Aviation Flight University of China,Guanghan 618307,China)

机构地区：[1]中国民用航空飞行学院空中交通管理学院,广汉618307

出　　处：《航空工程进展》2024年第2期49-55,共7页Advances in Aeronautical Science and Engineering

摘　　要：现阶段利用陆空通话语音对管制员疲劳状态的研究中,大多只考虑了语音在时域或频域的变化,而忽视了疲劳会同时在时域与频域上产生影响。将三种疲劳状态下的陆空通话语音分别转化为可同时反映时域与频域特性的语音频谱图像,利用灰度共生矩阵提取四维典型的特征参数,对比管制员在不同状态下特征参数的变化情况,构建管制员疲劳检测模型并对输入特征进行检测。结果表明:利用语谱图特征结合传统特征作为输入特征的检测准确率最高,达到95.49%,较单一使用传统特征的检测准确率高出4%;管制员疲劳状态的变化会直观地反映在语谱图上,会对其特征值产生影响,利用这种影响对管制员疲劳状态进行检测,可以得到良好的检测结果。In the current research on the fatigue state of controllers using radiotelephony communication,most of them only consider the changes of voice in the time domain or frequency domain,while ignore that fatigue will af-fect the time domain and frequency domain at the same time.The voice of radiotelephony communication in the three fatigue states is converted into speech spectrum images that can reflect the characteristics of both the time do-main and frequency domain,and the grayscale co-existence matrix is used to extract the typical feature parameters in four dimensions.The changes of the characteristic parameters of the controllers in different states are compared,and the controller fatigue detection model is constructed to detect the input features.The results show that the detec-tion accuracy of using the spectrogram features combined with the traditional features as the input features is the highest(95.49%),which is 4%higher than that of the traditional features alone.The change of controllers fatigue state can intuitively reflected on the spectrogram and haves an impact on its eigenvalues,and the better detection re-sults can be obtained by using this influence to detect the controllers fatigue state.

关 键 词：管制员 疲劳检测 语谱图 灰度共生矩阵 机器学习 

分 类 号：V355[航空宇航科学与技术—人机与环境工程]