计算机工程与科学2024，Vol.46Issue(8)：1425-1432,8.DOI:10.3969/j.issn.1007-130X.2024.08.011
基于加权非负矩阵分解的异常声音检测方法研究
An abnormal sound detection method based on weighted non-negative matrix decomposition
潘雨青 1于浩 1李峰1
作者信息
- 1. 江苏大学计算机科学与通信工程学院,江苏 镇江 212013
- 折叠
摘要
Abstract
Existing abnormal sound detection methods often rely on strongly labeled data for train-ing,but high-quality strongly labeled audio data is difficult to annotate and costly to collect.Addressing the issues of poor training results and low accuracy caused by interference from non-stationary and time-varying noise when using weakly labeled data in current abnormal audio detection methods,a weighted non-negative matrix factorization(WNMF)method based on audio spectrum is proposed.This method utilizes WNMF to label weakly labeled and unlabeled data,and separates target sound events from back-ground noise.Under appropriate weight values,WNMF alters the importance of audio information in different frequency bands during labeling to suppress noise and improve separation quality,approaching the effect of fully supervised model training.Then,a convolutional neural network is used to generate frame-level predictions and audio label predictions.Simulation experiments show that this method im-proves the accuracy by 4.8％compared to traditional NMF methods for processing weakly labeled data.关键词
异常声音检测/弱标签和无标签数据/加权非负矩阵分解/卷积神经网络Key words
abnormal sound detection/weakly labeled and unlabeled data/weighted non-negative ma-trix factorization/convolutional neural networks分类
计算机与自动化引用本文复制引用
潘雨青,于浩,李峰..基于加权非负矩阵分解的异常声音检测方法研究[J].计算机工程与科学,2024,46(8):1425-1432,8.
