Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530192
G. Gruber, M. Neumayer, T. Bretterklieber, H. Wegleiter
For small engines in non-automotive powertrains the emissions share is already limited. The introduction and integration of ECU -systems for engine control and dedicated sensors in small engines are required. The ion current sensing technology could be a key enabler for next generation combustion diagnoses and maintenance of small engines. It is an add-on sensing system and aims on gaining knowledge about the combustion process from the ion current signal. In this paper we present an analysis of an ion current sensing system from a metrological point of view. We investigate the impact of the ignition system and the ion current sensing system on the ion current signal and calculate a measurement error. We present a potential parameter to characterize the combustion process independently of the instrumentation. The analysis represents a first approach on how to design robust ion current based ECU control systems.
{"title":"Metrological Analysis of an Ion Current Measurement System","authors":"G. Gruber, M. Neumayer, T. Bretterklieber, H. Wegleiter","doi":"10.1109/SAS51076.2021.9530192","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530192","url":null,"abstract":"For small engines in non-automotive powertrains the emissions share is already limited. The introduction and integration of ECU -systems for engine control and dedicated sensors in small engines are required. The ion current sensing technology could be a key enabler for next generation combustion diagnoses and maintenance of small engines. It is an add-on sensing system and aims on gaining knowledge about the combustion process from the ion current signal. In this paper we present an analysis of an ion current sensing system from a metrological point of view. We investigate the impact of the ignition system and the ion current sensing system on the ion current signal and calculate a measurement error. We present a potential parameter to characterize the combustion process independently of the instrumentation. The analysis represents a first approach on how to design robust ion current based ECU control systems.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132920040","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530040
Chih-Chung Yang, Yu-Ting Li, D. Chiang, P. Chiu, Yi-Cheng Lin, W. Hsiao
The oil quality after the long heating time is required to be examined frequently because the degradation of oils can be detrimental to human health. Several sensing methods have addressed oil degradation problems but currently there is no known techniques to solve the problem in both efficient and economical ways. Three sensing methods, i.e. an interdigital planar sensor integrated with a LCR meter, spectrophotometer method and tested sensing paper, are proposed to characterize the quality of two kinds of edible oils. It is found that the logarithm of impedance of oils is linearly related to the logarithm of measured frequency, implying that the oils are dielectric materials. The impedances of oils decrease linearly with the increase of heated duration and the capacitance ratio of oils is weakly dependent on the heating duration. The wavelengths of starting transmittance are significantly red-shifted as observed by a spectrophotometer when the oils are heated for a long time. The absorbance of the oils increases exponentially with the heating time. The tested paper indicates that the color change can exhibit a quick oil qualitative measurement within a few minutes, but lacks quantitative information. Each sensing method has different sampling time, precision and accuracy for measuring the oil degradation, and the sensing methods should be chosen according to the required needs.
{"title":"Comparison of Sensing Methods for Characterization of Heated Oils Degradation","authors":"Chih-Chung Yang, Yu-Ting Li, D. Chiang, P. Chiu, Yi-Cheng Lin, W. Hsiao","doi":"10.1109/SAS51076.2021.9530040","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530040","url":null,"abstract":"The oil quality after the long heating time is required to be examined frequently because the degradation of oils can be detrimental to human health. Several sensing methods have addressed oil degradation problems but currently there is no known techniques to solve the problem in both efficient and economical ways. Three sensing methods, i.e. an interdigital planar sensor integrated with a LCR meter, spectrophotometer method and tested sensing paper, are proposed to characterize the quality of two kinds of edible oils. It is found that the logarithm of impedance of oils is linearly related to the logarithm of measured frequency, implying that the oils are dielectric materials. The impedances of oils decrease linearly with the increase of heated duration and the capacitance ratio of oils is weakly dependent on the heating duration. The wavelengths of starting transmittance are significantly red-shifted as observed by a spectrophotometer when the oils are heated for a long time. The absorbance of the oils increases exponentially with the heating time. The tested paper indicates that the color change can exhibit a quick oil qualitative measurement within a few minutes, but lacks quantitative information. Each sensing method has different sampling time, precision and accuracy for measuring the oil degradation, and the sensing methods should be chosen according to the required needs.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115581415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530113
Bruce Wallace, S. Gagnon, A. Stinchcombe, Stephanie Yamin, R. Goubran, F. Knoefel
Aging related changes and pathology affecting cognition and the ability to drive are significant issues for individuals, their families and the general population. Ensuring that unsafe drivers have their license suspended or get the additional training they need is important for the safety of the general population. On the other hand, allowing a person to continue to drive as long as they are safe is important for the social, emotional and cognitive wellbeing of the individual. This paper presents results of a preliminary study to see if an automated assessment based on trained machine learning models can correctly classify simulator drives as safe or unsafe in comparison to expert driver assessment opinion. The results show that the machine learning is able to achieve 85% accuracy in comparison to the experts for a combined group of 47 drivers that included 20 Healthy Controls, 9 diagnosed with Lewy Body Dementia and 18 diagnosed with mild Dementia of Alzheimer's Type. This work shows the potential for automated driver simulation assessment, which could reduce the burden on clinicians regarding driver safety evaluation.
{"title":"Preliminary Results for the Automated Assessment of Driving Simulation Results for Drivers with Cognitive Decline","authors":"Bruce Wallace, S. Gagnon, A. Stinchcombe, Stephanie Yamin, R. Goubran, F. Knoefel","doi":"10.1109/SAS51076.2021.9530113","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530113","url":null,"abstract":"Aging related changes and pathology affecting cognition and the ability to drive are significant issues for individuals, their families and the general population. Ensuring that unsafe drivers have their license suspended or get the additional training they need is important for the safety of the general population. On the other hand, allowing a person to continue to drive as long as they are safe is important for the social, emotional and cognitive wellbeing of the individual. This paper presents results of a preliminary study to see if an automated assessment based on trained machine learning models can correctly classify simulator drives as safe or unsafe in comparison to expert driver assessment opinion. The results show that the machine learning is able to achieve 85% accuracy in comparison to the experts for a combined group of 47 drivers that included 20 Healthy Controls, 9 diagnosed with Lewy Body Dementia and 18 diagnosed with mild Dementia of Alzheimer's Type. This work shows the potential for automated driver simulation assessment, which could reduce the burden on clinicians regarding driver safety evaluation.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129476874","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530132
C. Liu, Jiang Liu, S. Shimamoto
The sign language recognition system plays an important role in the field of human-computer interaction. In the daily life of hearing-impaired people, sign language is used as the main tool to communicate with the world. Although sign language can satisfy simple conversation, it is difficult to deal with in some situations where a lot of conversation is required such as medical emergencies or educational consultation. This paper proposes a sign language recognition system based on Wi-Fi to improve the life of the disabled. The proposed system collects the Channel State Information (CSI) due to the change of hand movement. Through the analysis of all subcarriers, the amplitude of CSI is determined to reflect the characteristics of different sign languages, some high-frequency noise is removed in the amplitude of CSI to obtain a smoother signal Gesture feature. We propose a gesture feature extraction method based on the variance of time series and DTW algorithm is used to recognize nine common Japanese sign language gestures. We set two daily conditions to test the system, and the experimental results show that the system performs well in different conditions.
{"title":"Sign Language Estimation Scheme Employing Wi-Fi Signal","authors":"C. Liu, Jiang Liu, S. Shimamoto","doi":"10.1109/SAS51076.2021.9530132","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530132","url":null,"abstract":"The sign language recognition system plays an important role in the field of human-computer interaction. In the daily life of hearing-impaired people, sign language is used as the main tool to communicate with the world. Although sign language can satisfy simple conversation, it is difficult to deal with in some situations where a lot of conversation is required such as medical emergencies or educational consultation. This paper proposes a sign language recognition system based on Wi-Fi to improve the life of the disabled. The proposed system collects the Channel State Information (CSI) due to the change of hand movement. Through the analysis of all subcarriers, the amplitude of CSI is determined to reflect the characteristics of different sign languages, some high-frequency noise is removed in the amplitude of CSI to obtain a smoother signal Gesture feature. We propose a gesture feature extraction method based on the variance of time series and DTW algorithm is used to recognize nine common Japanese sign language gestures. We set two daily conditions to test the system, and the experimental results show that the system performs well in different conditions.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128660666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530158
Hari Krishna Salila Vijayalal Mohan, A. Malcolm
In a high-volume food and beverage production environment, non-destructive and real-time inspection of various stages of food production from raw content processing to product packaging at high speed is a challenge. Specifically, filling and dispensing, packaging, and sealing lines encounter issues such as powder caking, non-homogenous powder composition, misaligned caps, and leaks during package sealing, which are currently addressed using human inspection and/or destructive, expensive and offline screening methodologies. In this work, a non-destructive evaluation platform using a capacitive proximity sensor was proposed and demonstrated to showcase novel applications such as monitoring powder caking, non-invasive powder composition analysis, contactless capping closure integrity testing and non-contact leak detection in sachet seals with high throughput, in-line integration capability and a small system footprint.
{"title":"Non-Destructive Evaluation of Food and Beverage (F&B) Fast Moving Consumer Goods (FMCG) Using Capacitive Proximity Sensor","authors":"Hari Krishna Salila Vijayalal Mohan, A. Malcolm","doi":"10.1109/SAS51076.2021.9530158","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530158","url":null,"abstract":"In a high-volume food and beverage production environment, non-destructive and real-time inspection of various stages of food production from raw content processing to product packaging at high speed is a challenge. Specifically, filling and dispensing, packaging, and sealing lines encounter issues such as powder caking, non-homogenous powder composition, misaligned caps, and leaks during package sealing, which are currently addressed using human inspection and/or destructive, expensive and offline screening methodologies. In this work, a non-destructive evaluation platform using a capacitive proximity sensor was proposed and demonstrated to showcase novel applications such as monitoring powder caking, non-invasive powder composition analysis, contactless capping closure integrity testing and non-contact leak detection in sachet seals with high throughput, in-line integration capability and a small system footprint.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122306709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530102
Vivek Bharati
Autonomy of the blind and visually impaired can be achieved through technological means and thereby empowering them with a sense of independence. Mobile phones are ubiquitous and can access artificial intelligence capabilities locally and in the Cloud. Navigational sensors, such as Light Detection and Ranging (LiDAR), and wide angle cameras, typically found in self-driving cars, are beginning to be incorporated into mobile phones. In this paper, we propose techniques for using mobile phone LiDAR + camera sensor data fusion along with edge + Cloud split AI to create an indoor situational awareness and navigational aid for the visually impaired. In addition to physical sensors, the system uses AI models as virtual sensors to provide the required functionality. The system enhances the image of a scene captured by a camera using distance information from the LiDAR and directional information computed by the device to provide a rich 3-D description of the space in front of the user. The system also uses a combination of sensor data fusion and geometric formulas to provide step-by-step walking instructions for the user in order to reach destinations. The user-centric system proposed here can be a valuable assistive technology for the blind and visually imnpired.
{"title":"LiDAR + Camera Sensor Data Fusion On Mobiles With AI-based Virtual Sensors To Provide Situational Awareness For The Visually Impaired","authors":"Vivek Bharati","doi":"10.1109/SAS51076.2021.9530102","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530102","url":null,"abstract":"Autonomy of the blind and visually impaired can be achieved through technological means and thereby empowering them with a sense of independence. Mobile phones are ubiquitous and can access artificial intelligence capabilities locally and in the Cloud. Navigational sensors, such as Light Detection and Ranging (LiDAR), and wide angle cameras, typically found in self-driving cars, are beginning to be incorporated into mobile phones. In this paper, we propose techniques for using mobile phone LiDAR + camera sensor data fusion along with edge + Cloud split AI to create an indoor situational awareness and navigational aid for the visually impaired. In addition to physical sensors, the system uses AI models as virtual sensors to provide the required functionality. The system enhances the image of a scene captured by a camera using distance information from the LiDAR and directional information computed by the device to provide a rich 3-D description of the space in front of the user. The system also uses a combination of sensor data fusion and geometric formulas to provide step-by-step walking instructions for the user in order to reach destinations. The user-centric system proposed here can be a valuable assistive technology for the blind and visually imnpired.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122479360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530079
D. Spirjakin, A. Baranov, S. Akbari, C. T. Phong, N. N. Tuan
Catalytic gas sensors are among the most widespread gas sensors for combustible gas concentration measurements. However, their selectivity is low. In this research, the results of machine learning techniques application to enhance catalytic gas sensor selectivity are presented. The measurements of sensor signal are performed using the multistage heat pulse method described in our previous works. Contrary to the previous works, the number of heating stages was increased from 2 to 55, which corresponds to the heating voltage range of 125 m V to 1.5 V with a 25 m V step. This change enriches sensor signal with information about gas compositions. Methane and vapors of acetone, ethanol and gasoline are used as target gases. A support vector machine method is used to train two models. The first one was trained based on the plain normalized data. It was used for a microcontroller implementation of the method. The second model used the data transformed by principal component analysis technique. This model was used to visualize the method proposed. The results show that the application of proposed method allows to identify gases by single catalytic sensor. These principles can be used to design selective gas detectors which will react only to target gases.
催化气体传感器是可燃气体浓度测量中应用最广泛的气体传感器之一。然而，它们的选择性很低。在本研究中，介绍了机器学习技术在提高催化气体传感器选择性方面的应用结果。传感器信号的测量使用了我们之前工作中描述的多级热脉冲法。与之前的工作相反，加热阶段的数量从2个增加到55个，对应于加热电压范围为125 m V至1.5 V，步长为25 m V。这种变化丰富了传感器信号中的气体成分信息。甲烷和丙酮、乙醇和汽油的蒸气被用作目标气体。采用支持向量机方法对两个模型进行训练。第一个是基于普通归一化数据进行训练的。它是用单片机实现的一种方法。第二个模型使用主成分分析技术转换后的数据。该模型用于可视化所提出的方法。结果表明，该方法的应用可以实现单催化传感器对气体的识别。这些原理可用于设计只对目标气体起反应的选择性气体探测器。
{"title":"Novel Method of Temperature Modulation for Enhancing Catalytic Gas Sensor Selectivity","authors":"D. Spirjakin, A. Baranov, S. Akbari, C. T. Phong, N. N. Tuan","doi":"10.1109/SAS51076.2021.9530079","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530079","url":null,"abstract":"Catalytic gas sensors are among the most widespread gas sensors for combustible gas concentration measurements. However, their selectivity is low. In this research, the results of machine learning techniques application to enhance catalytic gas sensor selectivity are presented. The measurements of sensor signal are performed using the multistage heat pulse method described in our previous works. Contrary to the previous works, the number of heating stages was increased from 2 to 55, which corresponds to the heating voltage range of 125 m V to 1.5 V with a 25 m V step. This change enriches sensor signal with information about gas compositions. Methane and vapors of acetone, ethanol and gasoline are used as target gases. A support vector machine method is used to train two models. The first one was trained based on the plain normalized data. It was used for a microcontroller implementation of the method. The second model used the data transformed by principal component analysis technique. This model was used to visualize the method proposed. The results show that the application of proposed method allows to identify gases by single catalytic sensor. These principles can be used to design selective gas detectors which will react only to target gases.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133265702","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530046
Simon Nutsch, M. Sauer
In this paper a method to test the latency, accuracy and power as well as energy demand of carbon dioxide sensors with the target on Heating, Ventilation and Air Conditioning (HVAC) applications is presented. In 24 trials the CO2 concentration in a measurement chamber was increased from ambient air to 1860 parts per million (ppm) in four steps. The CO2 concentration in the chamber was measured by the Testo 480 Indoor Air Quality (IAQ) analyzer and nine different non-dispersive infrared (NDIR) CO2 sensors. Furthermore, the design and components of the measurement chamber and the system to read the sensor values and measure the power and energy demand of the sensors are described. Although the measured data do not allow a statement about the actual sensor accuracy due to the small sample size and the accuracy of the used reference analyzer it is possible to declare if a sensor suitable for the application in demand control ventilation systems. To determine the sensor latency a method to measure the time a sensor needs to settle in a specific bound is shown.
{"title":"Method to determine the suitability of non-dispersive infrared carbon dioxide sensor models in Heating, Ventilation and Air Conditioning systems","authors":"Simon Nutsch, M. Sauer","doi":"10.1109/SAS51076.2021.9530046","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530046","url":null,"abstract":"In this paper a method to test the latency, accuracy and power as well as energy demand of carbon dioxide sensors with the target on Heating, Ventilation and Air Conditioning (HVAC) applications is presented. In 24 trials the CO2 concentration in a measurement chamber was increased from ambient air to 1860 parts per million (ppm) in four steps. The CO2 concentration in the chamber was measured by the Testo 480 Indoor Air Quality (IAQ) analyzer and nine different non-dispersive infrared (NDIR) CO2 sensors. Furthermore, the design and components of the measurement chamber and the system to read the sensor values and measure the power and energy demand of the sensors are described. Although the measured data do not allow a statement about the actual sensor accuracy due to the small sample size and the accuracy of the used reference analyzer it is possible to declare if a sensor suitable for the application in demand control ventilation systems. To determine the sensor latency a method to measure the time a sensor needs to settle in a specific bound is shown.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126455723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530085
M. Mozafari, R. Goubran, J. Green
Stress is a psychological condition that affects daily life, and chronic stress can result in cardiovascular disease and reduced productivity. Mental stress can be induced when difficult and time-limited tasks are assigned. Several groups have studied the relationship between physiologic signals and a subject's stress level. Through machine learning and signal processing, stress level can be automatically inferred from raw physiologic signals. As each person can have a specific physiologic reaction pattern to stress, it becomes problematic for a classifier to work well on a new subject. In this study, transfer learning is used to solve the problem of inter-subject variability. Methods are developed here to classify five levels of stress based on physiologic signals comprising photoplethysmogram (PPG), galvanic skin response (GSR), abdominal respiration, and thoracic respiration. Domain adaptation methods based on information-theoretical learning and transfer component analysis (TCA) are shown to reduce inter-subject variability of both GSR and respiratory signals. A fusion model was also designed to combine classification scores from each signal to reduce the effect of low-quality recording. The proposed method is shown to increase accuracy from 68.79% to 76.70% and Intraclass Correlation Coefficient (ICC) from 83.82% to 96.55%.
{"title":"A Fusion Model for Cross-Subject Stress Level Detection Based on Transfer Learning","authors":"M. Mozafari, R. Goubran, J. Green","doi":"10.1109/SAS51076.2021.9530085","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530085","url":null,"abstract":"Stress is a psychological condition that affects daily life, and chronic stress can result in cardiovascular disease and reduced productivity. Mental stress can be induced when difficult and time-limited tasks are assigned. Several groups have studied the relationship between physiologic signals and a subject's stress level. Through machine learning and signal processing, stress level can be automatically inferred from raw physiologic signals. As each person can have a specific physiologic reaction pattern to stress, it becomes problematic for a classifier to work well on a new subject. In this study, transfer learning is used to solve the problem of inter-subject variability. Methods are developed here to classify five levels of stress based on physiologic signals comprising photoplethysmogram (PPG), galvanic skin response (GSR), abdominal respiration, and thoracic respiration. Domain adaptation methods based on information-theoretical learning and transfer component analysis (TCA) are shown to reduce inter-subject variability of both GSR and respiratory signals. A fusion model was also designed to combine classification scores from each signal to reduce the effect of low-quality recording. The proposed method is shown to increase accuracy from 68.79% to 76.70% and Intraclass Correlation Coefficient (ICC) from 83.82% to 96.55%.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128015361","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-23DOI: 10.1109/SAS51076.2021.9530145
Bruno Tardif, D. Lo, R. Goubran
Exposure to gunshot sounds can cause hearing impairments. Measuring and analyzing these sounds can improve the design of hearing protectors and can help in enacting safety regulations. Furthermore, analyzing gunshot sounds can help identify the type of gun used. This is important for determining the appropriate public safety actions when a gunshot sound is detected in a public space. In this paper, we collected acoustic data from four different guns. To capture their sound including any non-symmetric sound propagation, 27 high dynamic range pressure microphones were placed around the guns forming a polar grid pattern. Audio signals were captured at 204.8 kHz sampling rate synchronously to preserve the fidelity of the impulse nature of the gunshots. In this study, an image-based analysis method was developed to take advantage of the recent advancement of image recognition techniques. Two spectral analysis methods: Short Time Fourier Transform (STFT) or Continuous Wavelet Transform (CWT), were then applied to get the spectrogram of the gunshot audio signal. Machine learning using the k-nearest neighbor and random subspaces was used to classify these spectrograms and identify which gun did the particular gunshot originated from. Under reverberant conditions, the STFT maintained a better identification accuracy than the CWT.
{"title":"Gunshot Sound Measurement and Analysis","authors":"Bruno Tardif, D. Lo, R. Goubran","doi":"10.1109/SAS51076.2021.9530145","DOIUrl":"https://doi.org/10.1109/SAS51076.2021.9530145","url":null,"abstract":"Exposure to gunshot sounds can cause hearing impairments. Measuring and analyzing these sounds can improve the design of hearing protectors and can help in enacting safety regulations. Furthermore, analyzing gunshot sounds can help identify the type of gun used. This is important for determining the appropriate public safety actions when a gunshot sound is detected in a public space. In this paper, we collected acoustic data from four different guns. To capture their sound including any non-symmetric sound propagation, 27 high dynamic range pressure microphones were placed around the guns forming a polar grid pattern. Audio signals were captured at 204.8 kHz sampling rate synchronously to preserve the fidelity of the impulse nature of the gunshots. In this study, an image-based analysis method was developed to take advantage of the recent advancement of image recognition techniques. Two spectral analysis methods: Short Time Fourier Transform (STFT) or Continuous Wavelet Transform (CWT), were then applied to get the spectrogram of the gunshot audio signal. Machine learning using the k-nearest neighbor and random subspaces was used to classify these spectrograms and identify which gun did the particular gunshot originated from. Under reverberant conditions, the STFT maintained a better identification accuracy than the CWT.","PeriodicalId":224327,"journal":{"name":"2021 IEEE Sensors Applications Symposium (SAS)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122007627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：