基于音频识别的自动门控制器设计  

Design of automatic door controller based on audio recognition

作　　者：陈波[1] 王鑫 李琦[1] CHEN Bo;WANG Xin;LI Qi(College of Information Engineer,Inner Mongolia University of Science and Technology,Baotou 014010,China)

机构地区：[1]内蒙古科技大学信息工程学院,内蒙古包头014010

出　　处：《现代电子技术》2021年第16期70-74,共5页Modern Electronics Technique

基　　金：内蒙古自治区科技重大专项(2019ZD025);内蒙古自治区科技成果转化项目(CGZH2018041);内蒙古自治区高等学校科学研究项目(NJZY21400)。

摘　　要：为解决草场车辆通行不便、传统围栏门效率低下等问题,在草场这种特殊环境下提出一种基于音频识别的自动门控制技术。该设计以STM32F407作为硬件核心,采用RT-Thread嵌入式实时操作系统。根据草场的环境复杂程度采集各种音频信号,采用快速傅里叶变换将音频信号变换至频域,主要针对车辆鸣笛声与其他干扰声音在频域的差别进行研究和分析,通过研究发现车辆的频谱具有多峰和倍频特性。根据这种特性设计音频识别算法,从而实现车辆识别。结果表明,相对于其他识别车辆方法,所提技术在成本和环保方面均更优,而且具有较高的识别精度和具有更加广阔的发展前景。In order to solve the problems of the inconvenience of vehicle traffic in pasture and inefficient passage of traditional artificial fence doors,an automatic door control technology based on audio recognition is proposed for the special environment of pasture.In this design,STM32F407ZET6 is taken as the hardware core and RT⁃Thread embedded real⁃time operating system is adopted.Firstly,all kinds of audio signals are collected according to the complexity of grassland environment.The fast Fourier transform(FFT)is used to transform the audio signal to the frequency domain.The difference between the vehicle whistle sound and other interference sounds in the frequency domain is studied and analyzed in this paper.It is found in the study that the spectrum of vehicles has the multi⁃peak and double frequency characteristics.According to the characteristics,an audio recognition algorithm is designed to realize vehicle recognition.The results show that,in comparison with other methods for vehicle identification,the proposed technology is better in terms of cost and environmental protection,and has higher recognition accuracy and broader development prospect.

关 键 词：自动门控制 音频识别 信号采集 车辆识别 音频信号变换 模拟仿真 

分 类 号：TN911.23-34[电子电信—通信与信息系统]