基于人工智能的噪声源识别技术研究  

Research on Noise Source Identification Technology Based on Artificial Intelligence

作　　者：冯继克 班倩茹 FENG Jike;BAN Qianru(Zhengzhou Institute of Technology and Business,Zhengzhou 450000,China)

机构地区：[1]郑州工商学院,河南郑州450000

出　　处：《电声技术》2025年第1期47-49,共3页Audio Engineering

摘　　要：为研究基于人工智能的噪声源识别技术,结合深度学习方法,提出一种从噪声数据中自动提取声学特征、分类噪声源并精确定位的方法。通过采用卷积神经网络、深度置信网络(Deep Belief Network,DBN)以及卷积-胶囊网络等技术,实现了高精度的噪声源分类与定位。实验表明,相比于传统方法,该方法在识别准确率、定位精度及计算效率方面有明显提升。In order to study the noise source identification technology based on artificial intelligence,a method is proposed to automatically extract acoustic features from noise data,classify noise sources,and accurately locate them by using deep learning techniques.The approach uses convolutional neural networks,Deep Belief Network(DBN),and convolutional capsule networks.Experiments show that this method improves recognition accuracy,localization precision,and computational efficiency compared to traditional methods.

关 键 词：人工智能 噪声源识别 深度学习 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]