The various bioacoustics signals obtained with auscultation contain complex clinical information that has been traditionally used as biomarkers,however,they are not extensively used in clinical studies owing to their ...The various bioacoustics signals obtained with auscultation contain complex clinical information that has been traditionally used as biomarkers,however,they are not extensively used in clinical studies owing to their spatiotemporal limitations.In this study,we developed a wearable stethoscope for wireless,skinattachable,low-power,continuous,real-time auscultation using a lung-sound-monitoring-patch(LSMP).LSMP can monitor respiratory function through a mobile app and classify normal and adventitious breathing by comparing their unique acoustic characteristics.The human heart and breathing sounds from humans can be distinguished from complex sound signals consisting of a mixture of bioacoustic signals and external noise.The performance of the LSMP sensor was further demonstrated in pediatric patients with asthma and elderly chronic obstructive pulmonary disease(COPD)patients where wheezing sounds were classified at specific frequencies.In addition,we developed a novel method for counting wheezing events based on a two-dimensional convolutional neural network deep-learning model constructed de novo and trained with our augmented fundamental lung-sound data set.We implemented a counting algorithm to identify wheezing events in real-time regardless of the respiratory cycle.The artificial intelligence-based adventitious breathing event counter distinguished>80%of the events(especially wheezing)in long-term clinical applications in patients with COPD.展开更多
AIM In our previous study, we have built a nine-gene(GPC3, HGF, ANXA1, FOS, SPAG9, HSPA1 B, CXCR4, PFN1, and CALR) expression detection system based on the Ge XP system. Based on peripheral blood and Ge XP, we aimed t...AIM In our previous study, we have built a nine-gene(GPC3, HGF, ANXA1, FOS, SPAG9, HSPA1 B, CXCR4, PFN1, and CALR) expression detection system based on the Ge XP system. Based on peripheral blood and Ge XP, we aimed to analyze the results of genes expression by different multi-parameter analysis methods and build a diagnostic model to classify hepatocellular carcinoma(HCC) patients and healthy people.METHODS Logistic regression analysis, discriminant analysis, classification tree analysis, and artificial neural network were used for the multi-parameter gene expression analysis method. One hundred and three patients with early HCC and 54 age-matched healthy normal controls were used to build a diagnostic model. Fiftytwo patients with early HCC and 34 healthy people were used for validation. The area under the curve, sensitivity, and specificity were used as diagnostic indicators.RESULTS Artificial neural network of the total nine genes had the best diagnostic value, and the AUC, sensitivity, and specificity were 0.943, 98%, and 85%, respectively. At last, 52 HCC patients and 34 healthy normal controls were used for validation. The sensitivity and specificity were 96% and 86%, respectively.CONCLUSION Multi-parameter analysis methods may increase the diagnostic value compared to single factor analysis and they may be a trend of the clinical diagnosis in the future.展开更多
In this paper, we present an approach to improve the accuracy of environmental sound event detection in a wireless acoustic sensor network for home monitoring. Wireless acoustic sensor nodes can capture sounds in the ...In this paper, we present an approach to improve the accuracy of environmental sound event detection in a wireless acoustic sensor network for home monitoring. Wireless acoustic sensor nodes can capture sounds in the home and simultaneously deliver them to a sink node for sound event detection. The proposed approach is mainly composed of three modules, including signal estimation, reliable sensor channel selection, and sound event detection. During signal estimation, lost packets are recovered to improve the signal quality. Next, reliable channels are selected using a multi-channel cross-correlation coefficient to improve the computational efficiency for distant sound event detection without sacrificing performance. Finally, the signals of the selected two channels are used for environmental sound event detection based on bidirectional gated recurrent neural networks using two-channel audio features. Experiments show that the proposed approach achieves superior performances compared to the baseline.展开更多
In this paper, we demonstrate the prototype of a new stethoscope using laser technology to make the heart-beat signal “visible”. This heartbeat detection technique could overcome the limitation of the acoustic steth...In this paper, we demonstrate the prototype of a new stethoscope using laser technology to make the heart-beat signal “visible”. This heartbeat detection technique could overcome the limitation of the acoustic stethoscope brought by the poor ability of human ear to hear low frequency heart sounds. This is important, as valuable information from sub-audio sounds is present at frequencies below the range of human hearing. Moreover, the diagnostic accuracy of the acoustic stethoscope is also very sensitive to noise from immediate environment. In the prototype of laser-based stethoscope, the heartbeat signal is correlated to the optical spot of a laser beam reflected from a thin mirror attached to the patient’s chest skin. The motion of the mirror with the chest skin is generated by the heart sounds. A linear optical sensor is applied to detect and record the motion of the optical spot, from which the heartbeat signal in time-domain is extracted. The heartbeat signal is then transformed to frequency domain through digital signal processing. Both time-domain and frequency-domain signals are analyzed in order to classify different type of heart murmurs. In the prototype of the laser-based stethoscope, we use a heart-sound box to simulate the chest of a human being. The heart-sounds collected from real patients are applied to activate the vibration of the heart-sound box. We also analyze different heart murmur patterns based on the time-domain and frequency-domain heartbeat signals acquired from the optical system.展开更多
The performance of classic Mel-frequency cepstral coefficients (MFCC) is unsatisfactory in noisy environment with different sound sources from nature. In this paper, a classification approach of the ecological environ...The performance of classic Mel-frequency cepstral coefficients (MFCC) is unsatisfactory in noisy environment with different sound sources from nature. In this paper, a classification approach of the ecological environmental sounds using the double-level energy detection (DED) was presented. The DED was used to detect the existence of the sound signals under noise conditions. In addition, MFCC features from the frames which were detected the presence of the sound signals by DED were extracted. Experimental results show that the proposed technology has better noise immunity than classic MFCC, and also outperforms time-domain energy detection (TED) and frequency-domain energy detection (FED) respectively.展开更多
Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,...Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,sound event localization and detection(SELD)has become a very active research topic.This paper presents a deep learning-based multioverlapping sound event localization and detection algorithm in three-dimensional space.Log-Mel spectrum and generalized cross-correlation spectrum are joined together in channel dimension as input features.These features are classified and regressed in parallel after training by a neural network to obtain sound recognition and localization results respectively.The channel attention mechanism is also introduced in the network to selectively enhance the features containing essential information and suppress the useless features.Finally,a thourough comparison confirms the efficiency and effectiveness of the proposed SELD algorithm.Field experiments show that the proposed algorithm is robust to reverberation and environment and can achieve higher recognition and localization accuracy compared with the baseline method.展开更多
In both industrial and research areas of electronic engineering,Sound Source Localization for robot control has always been an interesting subject to be further studied.Under some dangerous situation,especially when a...In both industrial and research areas of electronic engineering,Sound Source Localization for robot control has always been an interesting subject to be further studied.Under some dangerous situation,especially when a special driver is required to implement a particular task,the device should be able to combine robotics control technology with Sound Source Localization,and take actions according to the different response patterns.In this research project,a multifunc-tional model driver,named "Mobile Island",has been designed and built up by integrating the Emulator 8051 micro-controller,Intel 8255 interfaces,some components and other necessary devices.The intelligent Mobile Island imple-mented by C language programs can operate under three control modes.In the sound control Mode 1,the model driver can detect and track a target by Sound Source Localization and then turn and move toward the destination.In the keypad control Mode 2,it can be controlled by a manual keypad.In the free run Mode 3,Mobile Island can move and turn by itself.When finding an object in front,it will turn away before moving forward again,so that it can avoid crashing on the obstacle.展开更多
An example of using ultrasonic method to detect the compactness of complicated concrete-filled steel tube in certain high-rise building was discussed in this study.Because of the particularity of the complicated concr...An example of using ultrasonic method to detect the compactness of complicated concrete-filled steel tube in certain high-rise building was discussed in this study.Because of the particularity of the complicated concrete-filled steel tubular column,the plane detection method and embedded sounding pipe method were adopted in the process of effectively detecting the column.According to the results of the plane detection method and embedded sounding pipe method,the cementing status of steel tube and concrete can be concluded,which cannot be judged by the hammering method in the rectangular steel tube-reinforced concrete.展开更多
Precipitation detection is an essential step in radiance assimilation because the uncertainties in precipitation would affect the radiative transfer calculation and observation errors.The traditional precipitation det...Precipitation detection is an essential step in radiance assimilation because the uncertainties in precipitation would affect the radiative transfer calculation and observation errors.The traditional precipitation detection method for microwave only detects clouds and precipitation horizontally,without considering the three-dimensional distribution of clouds.Extending precipitation detection from 2D to 3D is expected to bring more useful information to the data assimilation without using the all-sky approach.In this study,the 3D precipitation detection method is adopted to assimilate Microwave Temperature Sounder-2(MWTS-Ⅱ)onboard the Fengyun-3D,which can dynamically detect the channels above precipitating clouds by considering the near-real-time cloud parameters.Cycling data assimilation and forecasting experiments for Typhoons Lekima(2019)and Mitag(2019)are carried out.Compared with the control experiment,the quantity of assimilated data with the 3D precipitation detection increases by approximately 23%.The quality of the additional MWTS-Ⅱradiance data is close to the clear-sky data.The case studies show that the average root-mean-square errors(RMSE)of prognostic variables are reduced by 1.7%in the upper troposphere,leading to an average reduction of4.53%in typhoon track forecasts.The detailed diagnoses of Typhoon Lekima(2019)further show that the additional MWTS-Ⅱradiances brought by the 3D precipitation detection facilitate portraying a more reasonable circulation situation,thus providing more precise structures.This paper preliminarily proves that 3D precipitation detection has potential added value for increasing satellite data utilization and improving typhoon forecasts.展开更多
This paper presents a design of new type of multi-parameter wearable medical devices signal processing platform. The signal processing algorithm has a QRS-wave detection algorithm based on LADT, wavelet transformation...This paper presents a design of new type of multi-parameter wearable medical devices signal processing platform. The signal processing algorithm has a QRS-wave detection algorithm based on LADT, wavelet transformation and threshold detection with TMS320VC5509 DSP system. The DSP can greatly increase the speed of QRS-wave detection, and the results can be practical used for multi-parameter wearable device detection of abnormal ECG.展开更多
Bactrocera dorsalis Hendel （Diptera： Tephritidae） is an invasive pest around the world. The paper summarizes biological and ecological characteristics of B, dorsalis, and reviews its detection methods from the aspe...Bactrocera dorsalis Hendel （Diptera： Tephritidae） is an invasive pest around the world. The paper summarizes biological and ecological characteristics of B, dorsalis, and reviews its detection methods from the aspects of morphological identification, acoustic detection and molecular detection, in order to provide a reference for further research and development of new detection methods. The hot issues in the study of B. dorsalis, such as ecological adaptation pattern, diffusion pathways and mechanisms, sustainable control measures, are also put forward in the paper.展开更多
Differentiating between regular and abnormal noises in machine-generated sounds is a crucial but difficult problem.For accurate audio signal classification,suitable and efficient techniques are needed,particularly mac...Differentiating between regular and abnormal noises in machine-generated sounds is a crucial but difficult problem.For accurate audio signal classification,suitable and efficient techniques are needed,particularly machine learning approaches for automated classification.Due to the dynamic and diverse representative characteristics of audio data,the probability of achieving high classification accuracy is relatively low and requires further research efforts.This study proposes an ensemble model based on the LeNet and hierarchical attention mechanism(HAM)models with MFCC features to enhance the models’capacity to handle bias.Additionally,CNNs,bidirectional LSTM(BiLSTM),CRNN,LSTM,capsule network model(CNM),attention mechanism(AM),gated recurrent unit(GRU),ResNet,EfficientNet,and HAM models are implemented for performance comparison.Experiments involving the DCASE2020 dataset reveal that the proposed approach works better than the others,achieving an impressive 99.13%accuracy and 99.56%k-fold cross-validation accuracy.Comparison with state-of-the-art studies further validates this performance.The study’s findings highlight the potential of the proposed approach for accurate fault detection in vehicles,particularly involving the use of acoustic data.展开更多
在海洋资源勘测和环境监测活动中,常需要准确测量水下声速信息。利用TDC-GP22设计了一种低成本的小型化声速仪,针对飞行时间(Time of Flight,TOF)法测声速过程中存在的超声波衰减等问题,提出了一种基于首波脉冲宽度的可变阈值算法,根据...在海洋资源勘测和环境监测活动中,常需要准确测量水下声速信息。利用TDC-GP22设计了一种低成本的小型化声速仪,针对飞行时间(Time of Flight,TOF)法测声速过程中存在的超声波衰减等问题,提出了一种基于首波脉冲宽度的可变阈值算法,根据回波信号的首波脉冲宽度值动态调整检测阈值,提高了测量结果的稳定性。同时通过分析声速测量过程中的常见误差来源,在系统动态标定环节设计了一种基于RBF神经网络的温度补偿算法,校准了温度带来的换能器频率偏移以及声电延时等误差。在不同温盐条件下的声速测量结果表明该系统能以较低的成本获取高精度水下声速信息。展开更多
基金supported by the Korea Environment Industry&Technology Institute(KEITI)through Digital Infrastructure Building Project for Monitoring,Surveying and Evaluating the Environmental Health program,funded by the Korea Ministry of Environment(MOE)(2021003330008)supported by the KIST Internal program(2E32851)+1 种基金supported by the Korea Health Technology Research and Development(R&D)Project through the Korea Health Industry Development Institute(KHIDI)and Korea Dementia Research Center(KDRC),funded by the Ministry of Health&Welfare and Ministry of Science and ICT,Republic of Korea(HU20C0164)the Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(2022R1A6A3A01087298)。
文摘The various bioacoustics signals obtained with auscultation contain complex clinical information that has been traditionally used as biomarkers,however,they are not extensively used in clinical studies owing to their spatiotemporal limitations.In this study,we developed a wearable stethoscope for wireless,skinattachable,low-power,continuous,real-time auscultation using a lung-sound-monitoring-patch(LSMP).LSMP can monitor respiratory function through a mobile app and classify normal and adventitious breathing by comparing their unique acoustic characteristics.The human heart and breathing sounds from humans can be distinguished from complex sound signals consisting of a mixture of bioacoustic signals and external noise.The performance of the LSMP sensor was further demonstrated in pediatric patients with asthma and elderly chronic obstructive pulmonary disease(COPD)patients where wheezing sounds were classified at specific frequencies.In addition,we developed a novel method for counting wheezing events based on a two-dimensional convolutional neural network deep-learning model constructed de novo and trained with our augmented fundamental lung-sound data set.We implemented a counting algorithm to identify wheezing events in real-time regardless of the respiratory cycle.The artificial intelligence-based adventitious breathing event counter distinguished>80%of the events(especially wheezing)in long-term clinical applications in patients with COPD.
基金National Key R&D Program of China,No.2016YFC0106604National Natural Science Foundation of China,No.81471761 and No.81501568
文摘AIM In our previous study, we have built a nine-gene(GPC3, HGF, ANXA1, FOS, SPAG9, HSPA1 B, CXCR4, PFN1, and CALR) expression detection system based on the Ge XP system. Based on peripheral blood and Ge XP, we aimed to analyze the results of genes expression by different multi-parameter analysis methods and build a diagnostic model to classify hepatocellular carcinoma(HCC) patients and healthy people.METHODS Logistic regression analysis, discriminant analysis, classification tree analysis, and artificial neural network were used for the multi-parameter gene expression analysis method. One hundred and three patients with early HCC and 54 age-matched healthy normal controls were used to build a diagnostic model. Fiftytwo patients with early HCC and 34 healthy people were used for validation. The area under the curve, sensitivity, and specificity were used as diagnostic indicators.RESULTS Artificial neural network of the total nine genes had the best diagnostic value, and the AUC, sensitivity, and specificity were 0.943, 98%, and 85%, respectively. At last, 52 HCC patients and 34 healthy normal controls were used for validation. The sensitivity and specificity were 96% and 86%, respectively.CONCLUSION Multi-parameter analysis methods may increase the diagnostic value compared to single factor analysis and they may be a trend of the clinical diagnosis in the future.
基金supported by Basic Science Research Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education (NRF2015R1D1A1A01059804)the MSIP (Ministry of Science,ICT and Future Planning),Korea,under the ITRC(Information Technology Research Center) support program (IITP-2016-R2718-16-0011) supervised by the IITP(Institute for Information & communications Technology Promotion)the present Research has been conducted by the Research Grant of Kwangwoon University in 2017
文摘In this paper, we present an approach to improve the accuracy of environmental sound event detection in a wireless acoustic sensor network for home monitoring. Wireless acoustic sensor nodes can capture sounds in the home and simultaneously deliver them to a sink node for sound event detection. The proposed approach is mainly composed of three modules, including signal estimation, reliable sensor channel selection, and sound event detection. During signal estimation, lost packets are recovered to improve the signal quality. Next, reliable channels are selected using a multi-channel cross-correlation coefficient to improve the computational efficiency for distant sound event detection without sacrificing performance. Finally, the signals of the selected two channels are used for environmental sound event detection based on bidirectional gated recurrent neural networks using two-channel audio features. Experiments show that the proposed approach achieves superior performances compared to the baseline.
文摘In this paper, we demonstrate the prototype of a new stethoscope using laser technology to make the heart-beat signal “visible”. This heartbeat detection technique could overcome the limitation of the acoustic stethoscope brought by the poor ability of human ear to hear low frequency heart sounds. This is important, as valuable information from sub-audio sounds is present at frequencies below the range of human hearing. Moreover, the diagnostic accuracy of the acoustic stethoscope is also very sensitive to noise from immediate environment. In the prototype of laser-based stethoscope, the heartbeat signal is correlated to the optical spot of a laser beam reflected from a thin mirror attached to the patient’s chest skin. The motion of the mirror with the chest skin is generated by the heart sounds. A linear optical sensor is applied to detect and record the motion of the optical spot, from which the heartbeat signal in time-domain is extracted. The heartbeat signal is then transformed to frequency domain through digital signal processing. Both time-domain and frequency-domain signals are analyzed in order to classify different type of heart murmurs. In the prototype of the laser-based stethoscope, we use a heart-sound box to simulate the chest of a human being. The heart-sounds collected from real patients are applied to activate the vibration of the heart-sound box. We also analyze different heart murmur patterns based on the time-domain and frequency-domain heartbeat signals acquired from the optical system.
文摘The performance of classic Mel-frequency cepstral coefficients (MFCC) is unsatisfactory in noisy environment with different sound sources from nature. In this paper, a classification approach of the ecological environmental sounds using the double-level energy detection (DED) was presented. The DED was used to detect the existence of the sound signals under noise conditions. In addition, MFCC features from the frames which were detected the presence of the sound signals by DED were extracted. Experimental results show that the proposed technology has better noise immunity than classic MFCC, and also outperforms time-domain energy detection (TED) and frequency-domain energy detection (FED) respectively.
基金supported by the National Natural Science Foundation of China(61877067)the Foundation of Science and Technology on Near-Surface Detection Laboratory(TCGZ2019A002,TCGZ2021C003,6142414200511)the Natural Science Basic Research Program of Shaanxi(2021JZ-19)。
文摘Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,sound event localization and detection(SELD)has become a very active research topic.This paper presents a deep learning-based multioverlapping sound event localization and detection algorithm in three-dimensional space.Log-Mel spectrum and generalized cross-correlation spectrum are joined together in channel dimension as input features.These features are classified and regressed in parallel after training by a neural network to obtain sound recognition and localization results respectively.The channel attention mechanism is also introduced in the network to selectively enhance the features containing essential information and suppress the useless features.Finally,a thourough comparison confirms the efficiency and effectiveness of the proposed SELD algorithm.Field experiments show that the proposed algorithm is robust to reverberation and environment and can achieve higher recognition and localization accuracy compared with the baseline method.
基金This paper is an introduction of the Research Project of‘LEC254’,which is held by The Hong Kong University of Science and Technology (HKUST)
文摘In both industrial and research areas of electronic engineering,Sound Source Localization for robot control has always been an interesting subject to be further studied.Under some dangerous situation,especially when a special driver is required to implement a particular task,the device should be able to combine robotics control technology with Sound Source Localization,and take actions according to the different response patterns.In this research project,a multifunc-tional model driver,named "Mobile Island",has been designed and built up by integrating the Emulator 8051 micro-controller,Intel 8255 interfaces,some components and other necessary devices.The intelligent Mobile Island imple-mented by C language programs can operate under three control modes.In the sound control Mode 1,the model driver can detect and track a target by Sound Source Localization and then turn and move toward the destination.In the keypad control Mode 2,it can be controlled by a manual keypad.In the free run Mode 3,Mobile Island can move and turn by itself.When finding an object in front,it will turn away before moving forward again,so that it can avoid crashing on the obstacle.
文摘An example of using ultrasonic method to detect the compactness of complicated concrete-filled steel tube in certain high-rise building was discussed in this study.Because of the particularity of the complicated concrete-filled steel tubular column,the plane detection method and embedded sounding pipe method were adopted in the process of effectively detecting the column.According to the results of the plane detection method and embedded sounding pipe method,the cementing status of steel tube and concrete can be concluded,which cannot be judged by the hammering method in the rectangular steel tube-reinforced concrete.
基金jointly sponsored by the National Key Research and Development Program of China(Grant Nos.2018YFC1506701 and 2017YFC1502102)the National Natural Science Foundation of China(Grant No.41675102)。
文摘Precipitation detection is an essential step in radiance assimilation because the uncertainties in precipitation would affect the radiative transfer calculation and observation errors.The traditional precipitation detection method for microwave only detects clouds and precipitation horizontally,without considering the three-dimensional distribution of clouds.Extending precipitation detection from 2D to 3D is expected to bring more useful information to the data assimilation without using the all-sky approach.In this study,the 3D precipitation detection method is adopted to assimilate Microwave Temperature Sounder-2(MWTS-Ⅱ)onboard the Fengyun-3D,which can dynamically detect the channels above precipitating clouds by considering the near-real-time cloud parameters.Cycling data assimilation and forecasting experiments for Typhoons Lekima(2019)and Mitag(2019)are carried out.Compared with the control experiment,the quantity of assimilated data with the 3D precipitation detection increases by approximately 23%.The quality of the additional MWTS-Ⅱradiance data is close to the clear-sky data.The case studies show that the average root-mean-square errors(RMSE)of prognostic variables are reduced by 1.7%in the upper troposphere,leading to an average reduction of4.53%in typhoon track forecasts.The detailed diagnoses of Typhoon Lekima(2019)further show that the additional MWTS-Ⅱradiances brought by the 3D precipitation detection facilitate portraying a more reasonable circulation situation,thus providing more precise structures.This paper preliminarily proves that 3D precipitation detection has potential added value for increasing satellite data utilization and improving typhoon forecasts.
文摘This paper presents a design of new type of multi-parameter wearable medical devices signal processing platform. The signal processing algorithm has a QRS-wave detection algorithm based on LADT, wavelet transformation and threshold detection with TMS320VC5509 DSP system. The DSP can greatly increase the speed of QRS-wave detection, and the results can be practical used for multi-parameter wearable device detection of abnormal ECG.
基金Supported by International Cooperation Project of the Ministry of Science and Technology(2011DFB30040)Key Project of Science and Technology Development Fund of Guangxi Academy of Agricultural Sciences(2012JZ08)Scientific and Technological Projects of Nanning Municipal Science and Technology Bureau(20132308)
文摘Bactrocera dorsalis Hendel （Diptera： Tephritidae） is an invasive pest around the world. The paper summarizes biological and ecological characteristics of B, dorsalis, and reviews its detection methods from the aspects of morphological identification, acoustic detection and molecular detection, in order to provide a reference for further research and development of new detection methods. The hot issues in the study of B. dorsalis, such as ecological adaptation pattern, diffusion pathways and mechanisms, sustainable control measures, are also put forward in the paper.
基金funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R746),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia。
文摘Differentiating between regular and abnormal noises in machine-generated sounds is a crucial but difficult problem.For accurate audio signal classification,suitable and efficient techniques are needed,particularly machine learning approaches for automated classification.Due to the dynamic and diverse representative characteristics of audio data,the probability of achieving high classification accuracy is relatively low and requires further research efforts.This study proposes an ensemble model based on the LeNet and hierarchical attention mechanism(HAM)models with MFCC features to enhance the models’capacity to handle bias.Additionally,CNNs,bidirectional LSTM(BiLSTM),CRNN,LSTM,capsule network model(CNM),attention mechanism(AM),gated recurrent unit(GRU),ResNet,EfficientNet,and HAM models are implemented for performance comparison.Experiments involving the DCASE2020 dataset reveal that the proposed approach works better than the others,achieving an impressive 99.13%accuracy and 99.56%k-fold cross-validation accuracy.Comparison with state-of-the-art studies further validates this performance.The study’s findings highlight the potential of the proposed approach for accurate fault detection in vehicles,particularly involving the use of acoustic data.
文摘在海洋资源勘测和环境监测活动中,常需要准确测量水下声速信息。利用TDC-GP22设计了一种低成本的小型化声速仪,针对飞行时间(Time of Flight,TOF)法测声速过程中存在的超声波衰减等问题,提出了一种基于首波脉冲宽度的可变阈值算法,根据回波信号的首波脉冲宽度值动态调整检测阈值,提高了测量结果的稳定性。同时通过分析声速测量过程中的常见误差来源,在系统动态标定环节设计了一种基于RBF神经网络的温度补偿算法,校准了温度带来的换能器频率偏移以及声电延时等误差。在不同温盐条件下的声速测量结果表明该系统能以较低的成本获取高精度水下声速信息。
