Coughing is an obvious respiratory disease symptom,which affects the airways and lungs of pigs.In pig houses,continuous online monitoring of cough sounds can be used to build an intelligent alarm system for disease ea...Coughing is an obvious respiratory disease symptom,which affects the airways and lungs of pigs.In pig houses,continuous online monitoring of cough sounds can be used to build an intelligent alarm system for disease early detection.Owing to complicated interferences in piggery,recognition of pig cough sound becomes difficult.Although a lot of algorithms have been proposed to recognize the pig cough sounds,the recognition accuracy in field sit-uations still needs enhancement.The purpose of this research is to provide a highly accu-rate pig cough recognition method for the respiratory disease alarm system.We propose a classification algorithm based on the fine-tuned AlexNet model and feature of the spectro-gram.With the advantages of the convolutional neural network in image recognition,the sound signals are converted into spectrogram images for recognition,to enhance the accu-racy.We compare the proposed algorithm’s performance with the probabilistic neural net-work classifier and some existing algorithms.The results reveal that the proposed algorithm significantly outperforms the other algorithms-cough and overall recognition accuracies reach to 96.8%and 95.4%,respectively,with 96.2%F1-score achieved.展开更多
咳嗽是呼吸道疾病中一种常见的症状,基于模式识别算法可以对语音信号中咳嗽对象的频度和强度进行客观化分析,进而帮助临床咳嗽的诊断及病程跟踪.在临床录制的连续语音信号中检测出咳嗽对象是咳嗽诊断及分析的基础.本文将咳嗽检测视为模...咳嗽是呼吸道疾病中一种常见的症状,基于模式识别算法可以对语音信号中咳嗽对象的频度和强度进行客观化分析,进而帮助临床咳嗽的诊断及病程跟踪.在临床录制的连续语音信号中检测出咳嗽对象是咳嗽诊断及分析的基础.本文将咳嗽检测视为模式识别中的二分类问题,借助于分类器将咳嗽对象从背景信号中分离.在深入研究咳嗽频谱分布的基础上,提出一种新的基于高频子带的特征提取方法(High-frequency subband features method),在提取咳嗽信号特征之前,使用高频滤波器获取高频部分信号.在合成实验数据的过程中使用了不同的噪声类型和信噪比来组成不同的实验环境,并且在每种实验环境下对几种特征提取方法进行了评价与分析.实验结果表明,相比于常见的语音信号特征,结合基于高频子带特征的咳嗽检测方法在检测正确率等性能指标上有显著地提升.展开更多
基金This work was supported by the grant from the National Key Research and Development Program of China under Grant 2016YFD0700204-02the Earmarked Fund for China Agricul-ture Research System under Grant CARS-35+2 种基金the"Young Talents"Project of Northeast Agricultural University under Grant 17QC20the University Nursing Program for Young Scholars with Creative Talents in Heilongjiang Province under Grant UNPYSCT-2020092 and UNPYSCT-2018142and the Hei-longjiang Post-doctoral Subsidy Project of China under Grant LBH-Z17020.
文摘Coughing is an obvious respiratory disease symptom,which affects the airways and lungs of pigs.In pig houses,continuous online monitoring of cough sounds can be used to build an intelligent alarm system for disease early detection.Owing to complicated interferences in piggery,recognition of pig cough sound becomes difficult.Although a lot of algorithms have been proposed to recognize the pig cough sounds,the recognition accuracy in field sit-uations still needs enhancement.The purpose of this research is to provide a highly accu-rate pig cough recognition method for the respiratory disease alarm system.We propose a classification algorithm based on the fine-tuned AlexNet model and feature of the spectro-gram.With the advantages of the convolutional neural network in image recognition,the sound signals are converted into spectrogram images for recognition,to enhance the accu-racy.We compare the proposed algorithm’s performance with the probabilistic neural net-work classifier and some existing algorithms.The results reveal that the proposed algorithm significantly outperforms the other algorithms-cough and overall recognition accuracies reach to 96.8%and 95.4%,respectively,with 96.2%F1-score achieved.
文摘咳嗽是呼吸道疾病中一种常见的症状,基于模式识别算法可以对语音信号中咳嗽对象的频度和强度进行客观化分析,进而帮助临床咳嗽的诊断及病程跟踪.在临床录制的连续语音信号中检测出咳嗽对象是咳嗽诊断及分析的基础.本文将咳嗽检测视为模式识别中的二分类问题,借助于分类器将咳嗽对象从背景信号中分离.在深入研究咳嗽频谱分布的基础上,提出一种新的基于高频子带的特征提取方法(High-frequency subband features method),在提取咳嗽信号特征之前,使用高频滤波器获取高频部分信号.在合成实验数据的过程中使用了不同的噪声类型和信噪比来组成不同的实验环境,并且在每种实验环境下对几种特征提取方法进行了评价与分析.实验结果表明,相比于常见的语音信号特征,结合基于高频子带特征的咳嗽检测方法在检测正确率等性能指标上有显著地提升.
