Pub Date : 2020-12-02DOI: 10.1109/ICMLC51923.2020.9469539
Yang Fei, Xin Yuan, P. Shi, C. Lim
For practical applications of multi-agent systems, agents could be continuously subject to external disturbances, which is likely to negatively affect their tracking performances. Furthermore, the unknown or inaccurate factors in system dynamics can also bring negative effects to a system’s performance. In this paper, an uncertainty-observer-based dynamic sliding mode control scheme is proposed to deal with time-varying formation control problems of second-order nonlinear multi-agent systems. An uncertainty observer is first implemented for each agent to estimate the combination of the agent’s external and internal uncertainties and its time derivative. An observer-based dynamic sliding mode formation control law is developed for a cluster of second-order nonlinear agents to achieve time-varying formation. Finally, a numerical simulation is provided to illustrate the validity of the proposed control approach.
{"title":"Uncertainty-Observer-Based Dynamic Sliding Mode Formation Control for Multi-Agent Systems","authors":"Yang Fei, Xin Yuan, P. Shi, C. Lim","doi":"10.1109/ICMLC51923.2020.9469539","DOIUrl":"https://doi.org/10.1109/ICMLC51923.2020.9469539","url":null,"abstract":"For practical applications of multi-agent systems, agents could be continuously subject to external disturbances, which is likely to negatively affect their tracking performances. Furthermore, the unknown or inaccurate factors in system dynamics can also bring negative effects to a system’s performance. In this paper, an uncertainty-observer-based dynamic sliding mode control scheme is proposed to deal with time-varying formation control problems of second-order nonlinear multi-agent systems. An uncertainty observer is first implemented for each agent to estimate the combination of the agent’s external and internal uncertainties and its time derivative. An observer-based dynamic sliding mode formation control law is developed for a cluster of second-order nonlinear agents to achieve time-varying formation. Finally, a numerical simulation is provided to illustrate the validity of the proposed control approach.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131631017","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 : 2020-12-02DOI: 10.1109/ICMLC51923.2020.9469592
Jingjing Song, Zehua Jiang, Huili Dou, Eric C. C. Tsang
In neighborhood based attribute reduction, neighborhood relation is a typical tool for distinguishing samples. Notably, the neighborhood relation may be powerless in providing satisfactory distinguishing ability. In view of this, the supervised neighborhood based attribute reduction has been explored. However, the supervised neighborhood based reduct may be lack of universality. To file such gap, an ensemble strategy for computing supervised neighborhood based reduct is proposed in our paper. Such ensemble strategy is realized through considering the requirement of each decision class. The experimental results on 8 UCI data sets show that the supervised neighborhood based ensemble strategy can generate reduct not only with higher generalization performance but also with higher stability.
{"title":"Supervised Neighborhood Based Ensemble Attribute Reduction","authors":"Jingjing Song, Zehua Jiang, Huili Dou, Eric C. C. Tsang","doi":"10.1109/ICMLC51923.2020.9469592","DOIUrl":"https://doi.org/10.1109/ICMLC51923.2020.9469592","url":null,"abstract":"In neighborhood based attribute reduction, neighborhood relation is a typical tool for distinguishing samples. Notably, the neighborhood relation may be powerless in providing satisfactory distinguishing ability. In view of this, the supervised neighborhood based attribute reduction has been explored. However, the supervised neighborhood based reduct may be lack of universality. To file such gap, an ensemble strategy for computing supervised neighborhood based reduct is proposed in our paper. Such ensemble strategy is realized through considering the requirement of each decision class. The experimental results on 8 UCI data sets show that the supervised neighborhood based ensemble strategy can generate reduct not only with higher generalization performance but also with higher stability.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123829914","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 : 2020-12-02DOI: 10.1109/ICMLC51923.2020.9469584
Michael Willis, Li Zhang, Han Liu, Hailun Xie, Kamlesh Mistry
The identification of the most discriminative features in an explainable AI decision-making process is a challenging problem. This research tackles such challenges by proposing Particle Swarm Optimization (PSO) variants embedded with novel mutation and sampling iteration operations for feature selection in object recognition. Specifically, five PSO variants integrating different mutation and sampling strategies have been proposed to select the most discriminative feature subsets for the classification of different objects. A mutation strategy is firstly proposed by randomly flipping the particle positions in some dimensions to generate new feature interactions. Moreover, instead of embarking the position updating evolution in PSO, the proposed PSO variants generate offspring solutions through a sampling mechanism during the initial search process. Two offspring generation sampling schemes are investigated, i.e. the employment of the personal and global best solutions obtained using the mutation mechanism, respectively, as the starting positions for the subsequent search process. Subsequently, several machine learning algorithms are used in conjunction with the proposed PSO variants to perform object classification. As evidenced by the empirical results, the proposed PSO variants outperform the original PSO algorithm, significantly, for feature optimization.
{"title":"Object Recognition Using Enhanced Particle Swarm Optimization","authors":"Michael Willis, Li Zhang, Han Liu, Hailun Xie, Kamlesh Mistry","doi":"10.1109/ICMLC51923.2020.9469584","DOIUrl":"https://doi.org/10.1109/ICMLC51923.2020.9469584","url":null,"abstract":"The identification of the most discriminative features in an explainable AI decision-making process is a challenging problem. This research tackles such challenges by proposing Particle Swarm Optimization (PSO) variants embedded with novel mutation and sampling iteration operations for feature selection in object recognition. Specifically, five PSO variants integrating different mutation and sampling strategies have been proposed to select the most discriminative feature subsets for the classification of different objects. A mutation strategy is firstly proposed by randomly flipping the particle positions in some dimensions to generate new feature interactions. Moreover, instead of embarking the position updating evolution in PSO, the proposed PSO variants generate offspring solutions through a sampling mechanism during the initial search process. Two offspring generation sampling schemes are investigated, i.e. the employment of the personal and global best solutions obtained using the mutation mechanism, respectively, as the starting positions for the subsequent search process. Subsequently, several machine learning algorithms are used in conjunction with the proposed PSO variants to perform object classification. As evidenced by the empirical results, the proposed PSO variants outperform the original PSO algorithm, significantly, for feature optimization.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129919542","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 : 2020-12-02DOI: 10.1109/ICMLC51923.2020.9469530
Asaduzzaman Sajeeb, A. Sakib, Sanjida Ali Shushmita, S. Kabir, Md. Tanzim Reza, M. Parvez
Parkinson’s disease(PD) is a neurological condition that is dynamic and steadily influences the movement of the human body. PD influences the central apprehensive system which happens because of the hardship of dopaminergic neurons brought about in a neuro-degenerative incubation. The patients who have PD usually suffer from tremor, unyielding nature, postural shifts, and lessen in unconstrained advancements. There is no particular diagnosis process for PD. PD varies from one person to another person depending on the situation and the family history.Magnetic Resonance Imaging (MRI), Computed Tomography (CT), ultrasound of the brain, Positron Emission Tomography (PET) scans are common imaging tests to figure out this disease but these tests are not particularly effective. In this research, several tests are run on two types of data group - control and PD affected people. The dataset is collected from the Parkinson’s Progression Markers Initiative (PPMI) repository. Then MRI slices are processed from the selected data group into the CNN models. Three different Convolutional Neural Network (CNN) architectures are used in this work to extract features from the data group. The CNN models are InceptionV3, VGG16 and VGG19. These models are used in this research to compare and to get better accuracy. Among these models, VGG19 worked best in the dataset because the accuracy for VGG19 is 91.5% where VGG16 gives 88.5% and inceptionV3 gives 89.5% accuracy on detecting PD.
{"title":"Parkinson’s Disease Detection Using FMRI Images Leveraging Transfer Learning on Convolutional Neural Network","authors":"Asaduzzaman Sajeeb, A. Sakib, Sanjida Ali Shushmita, S. Kabir, Md. Tanzim Reza, M. Parvez","doi":"10.1109/ICMLC51923.2020.9469530","DOIUrl":"https://doi.org/10.1109/ICMLC51923.2020.9469530","url":null,"abstract":"Parkinson’s disease(PD) is a neurological condition that is dynamic and steadily influences the movement of the human body. PD influences the central apprehensive system which happens because of the hardship of dopaminergic neurons brought about in a neuro-degenerative incubation. The patients who have PD usually suffer from tremor, unyielding nature, postural shifts, and lessen in unconstrained advancements. There is no particular diagnosis process for PD. PD varies from one person to another person depending on the situation and the family history.Magnetic Resonance Imaging (MRI), Computed Tomography (CT), ultrasound of the brain, Positron Emission Tomography (PET) scans are common imaging tests to figure out this disease but these tests are not particularly effective. In this research, several tests are run on two types of data group - control and PD affected people. The dataset is collected from the Parkinson’s Progression Markers Initiative (PPMI) repository. Then MRI slices are processed from the selected data group into the CNN models. Three different Convolutional Neural Network (CNN) architectures are used in this work to extract features from the data group. The CNN models are InceptionV3, VGG16 and VGG19. These models are used in this research to compare and to get better accuracy. Among these models, VGG19 worked best in the dataset because the accuracy for VGG19 is 91.5% where VGG16 gives 88.5% and inceptionV3 gives 89.5% accuracy on detecting PD.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129844004","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 : 2020-10-28DOI: 10.1109/ICMLC51923.2020.9469038
Zarif Ahmed Chowdhury, Dewan Nahidul Alam, Md. Abu Fattah Hossain Bhuiyan Nahid, Md Anisur Rahman, M. Parvez
Over the centuries, human aimed to achieve the ability to understand the inner functions of the mind and brain. One of the techniques to understand such functions is the application of neurofeedback. Neurofeedback is the procedure which has an influence on physiological brain conditions that takes place by allowing self-regulation of brain activities. Several techniques have been used in the application of neurofeedback to improve different kinds of brain-related conditions including attention capacity and other disabilities. However, literature shows that there are still chances of further improvement in this field, since neurofeedback often causes complications anxiety, discontent and discomfort. Therefore in this paper, we proposed a method to detect modulated motor cortex using anodal and cathodal tDCS based neurofeedback to achieve a better result in the application of neurofeedback. The proposed method showed a higher percentage of accuracy (98.67%) for both anodal and cathodal using Electroencephalography(EEG) based neurofeedback data for twenty subjects. The accuracy of our proposed method is better than three other existing techniques on neurofeedback application. The experimental results demonstrate that our proposed method is suitable in the application of neurofeedback.
{"title":"Detection of Modulated Motor Cortex using Anodal and Cathodal TDCS based Neurofeedback","authors":"Zarif Ahmed Chowdhury, Dewan Nahidul Alam, Md. Abu Fattah Hossain Bhuiyan Nahid, Md Anisur Rahman, M. Parvez","doi":"10.1109/ICMLC51923.2020.9469038","DOIUrl":"https://doi.org/10.1109/ICMLC51923.2020.9469038","url":null,"abstract":"Over the centuries, human aimed to achieve the ability to understand the inner functions of the mind and brain. One of the techniques to understand such functions is the application of neurofeedback. Neurofeedback is the procedure which has an influence on physiological brain conditions that takes place by allowing self-regulation of brain activities. Several techniques have been used in the application of neurofeedback to improve different kinds of brain-related conditions including attention capacity and other disabilities. However, literature shows that there are still chances of further improvement in this field, since neurofeedback often causes complications anxiety, discontent and discomfort. Therefore in this paper, we proposed a method to detect modulated motor cortex using anodal and cathodal tDCS based neurofeedback to achieve a better result in the application of neurofeedback. The proposed method showed a higher percentage of accuracy (98.67%) for both anodal and cathodal using Electroencephalography(EEG) based neurofeedback data for twenty subjects. The accuracy of our proposed method is better than three other existing techniques on neurofeedback application. The experimental results demonstrate that our proposed method is suitable in the application of neurofeedback.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117202427","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 : 2020-08-25DOI: 10.36227/techrxiv.12846563.v1
Philip Easom, A. Bouridane, Feiyu Qiang, Li Zhang, Carolyn Downs, Richard M. Jiang
With an increasing amount of elderly people needing home care around the clock, care workers are not able to keep up with the demand of providing maximum support to those who require it. As medical costs of home care increase the quality is care suffering as a result of staff shortages, a solution is desperately needed to make the valuable care time of these workers more efficient. This paper proposes a system that is able to make use of the deep learning resources currently available to produce a base system that could provide a solution to many of the problems that care homes and staff face today. Transfer learning was conducted on a deep convolutional neural network to recognize common household objects was proposed. This system showed promising results with an accuracy, sensitivity and specificity of 90.6%, 0.90977 and 0.99668 respectively. Real-time applications were also considered, with the system achieving a maximum speed of 19.6 FPS on an MSI GTX 1060 GPU with 4GB of VRAM allocated.
{"title":"In-House Deep Environmental Sentience for Smart Homecare Solutions Toward Ageing Society","authors":"Philip Easom, A. Bouridane, Feiyu Qiang, Li Zhang, Carolyn Downs, Richard M. Jiang","doi":"10.36227/techrxiv.12846563.v1","DOIUrl":"https://doi.org/10.36227/techrxiv.12846563.v1","url":null,"abstract":"With an increasing amount of elderly people needing home care around the clock, care workers are not able to keep up with the demand of providing maximum support to those who require it. As medical costs of home care increase the quality is care suffering as a result of staff shortages, a solution is desperately needed to make the valuable care time of these workers more efficient. This paper proposes a system that is able to make use of the deep learning resources currently available to produce a base system that could provide a solution to many of the problems that care homes and staff face today. Transfer learning was conducted on a deep convolutional neural network to recognize common household objects was proposed. This system showed promising results with an accuracy, sensitivity and specificity of 90.6%, 0.90977 and 0.99668 respectively. Real-time applications were also considered, with the system achieving a maximum speed of 19.6 FPS on an MSI GTX 1060 GPU with 4GB of VRAM allocated.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"259 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116060603","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 : 2020-05-19DOI: 10.1109/ICMLC51923.2020.9469565
Afsara Mashiat, Reza Rifat Akhlaque, Fahmeda Hasan Fariha, Md. Tanzim Reza, Md Anisur Rahman, M. Parvez
As brain is the most vital organ of the human body, the affects of brain related diseases can be severe. One of the most harmful diseases is brain tumor, which results in a very short life expectancy of the affected patient. Detection of brain tumor is a challenging task in the early stages. Still, with the help of modern technology and machine learning algorithms, it has become a matter of great interest for research. While detecting the brain tumor of an affected person, we are considering the fMRI data of the patient. Our aim is to identify whether the tumor is present in the patient’s brain or not. We use a Convolutional Neural Network(CNN) that is good enough to generate high accuracy. We have used some deeper architecture design VGG16, VGG19, and Inception v3 for better accuracy. Three classification techniques are used namely binary classification, lobe based classification, and position based classification. The main contribution of our proposed work is that we have identified the specific region of the brain where the tumor is located. The region-based classification distinguishes our work from others that are applied on the same dataset. For binary classification, we found approximately 95% accuracy from all the three architectures. Furthermore, we found approximately 78% accuracy for lobe based classification and approximately 97% accuracy for position based classification. The experimental results indicate the superiority of our proposed method in terms of identifying the brain tumor.
{"title":"Detection of Brain Tumor and Identification of Tumor Region Using Deep Neural Network On FMRI Images","authors":"Afsara Mashiat, Reza Rifat Akhlaque, Fahmeda Hasan Fariha, Md. Tanzim Reza, Md Anisur Rahman, M. Parvez","doi":"10.1109/ICMLC51923.2020.9469565","DOIUrl":"https://doi.org/10.1109/ICMLC51923.2020.9469565","url":null,"abstract":"As brain is the most vital organ of the human body, the affects of brain related diseases can be severe. One of the most harmful diseases is brain tumor, which results in a very short life expectancy of the affected patient. Detection of brain tumor is a challenging task in the early stages. Still, with the help of modern technology and machine learning algorithms, it has become a matter of great interest for research. While detecting the brain tumor of an affected person, we are considering the fMRI data of the patient. Our aim is to identify whether the tumor is present in the patient’s brain or not. We use a Convolutional Neural Network(CNN) that is good enough to generate high accuracy. We have used some deeper architecture design VGG16, VGG19, and Inception v3 for better accuracy. Three classification techniques are used namely binary classification, lobe based classification, and position based classification. The main contribution of our proposed work is that we have identified the specific region of the brain where the tumor is located. The region-based classification distinguishes our work from others that are applied on the same dataset. For binary classification, we found approximately 95% accuracy from all the three architectures. Furthermore, we found approximately 78% accuracy for lobe based classification and approximately 97% accuracy for position based classification. The experimental results indicate the superiority of our proposed method in terms of identifying the brain tumor.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129531438","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 : 2020-05-19DOI: 10.1109/ICMLC51923.2020.9469562
S. Sultana, Md Anisur Rahman, M. Parvez
Stress refers to body’s physical, emotional and psychological reaction to any environmental change needing adjustment with major impact on human psychology. Stress is specially difficult to manage for visually impaired people (VIP) as they can become easily stressed in unknown situations. Electroencephalogram (EEG) signals can be used to detect stress as it basically represents the ongoing electrical signal changes in human brain. Literature shows that the stress detection techniques are mostly based on either time or frequency domain analysis. However, using either time or frequency domain analysis may not be sufficient to provide appropriate outcome for stress detection. Hence, in this paper a method is proposed using empirical mode decomposition (EMD) and short-term Fourier transform (STFT) are used to extract features considering spatio-temporal information from EEG signals. In the EMD, the signal is first decomposed into intrinsic mode functions (IMFs) representing a finite number of signals while maintaining the time domain and STFT is used to convert time domain to time-frequency domain. Support vector machine (SVM) is applied to classify the stress of VIP in unfamiliar indoor environments. The performance of the proposed method is compared with a state-of-the-art technique for stress detection. The experimental results demonstrate the superiority of the proposed technique over the existing technique.
{"title":"Detection of Stress for Visually Impaired People Using EEG Signals Based on Time-Frequency Domain Analysis","authors":"S. Sultana, Md Anisur Rahman, M. Parvez","doi":"10.1109/ICMLC51923.2020.9469562","DOIUrl":"https://doi.org/10.1109/ICMLC51923.2020.9469562","url":null,"abstract":"Stress refers to body’s physical, emotional and psychological reaction to any environmental change needing adjustment with major impact on human psychology. Stress is specially difficult to manage for visually impaired people (VIP) as they can become easily stressed in unknown situations. Electroencephalogram (EEG) signals can be used to detect stress as it basically represents the ongoing electrical signal changes in human brain. Literature shows that the stress detection techniques are mostly based on either time or frequency domain analysis. However, using either time or frequency domain analysis may not be sufficient to provide appropriate outcome for stress detection. Hence, in this paper a method is proposed using empirical mode decomposition (EMD) and short-term Fourier transform (STFT) are used to extract features considering spatio-temporal information from EEG signals. In the EMD, the signal is first decomposed into intrinsic mode functions (IMFs) representing a finite number of signals while maintaining the time domain and STFT is used to convert time domain to time-frequency domain. Support vector machine (SVM) is applied to classify the stress of VIP in unfamiliar indoor environments. The performance of the proposed method is compared with a state-of-the-art technique for stress detection. The experimental results demonstrate the superiority of the proposed technique over the existing technique.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121372816","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 : 2020-05-16DOI: 10.1109/ICMLC51923.2020.9469537
Fraser Young, Li Zhang, Richard Jiang, Han Liu, Conor Wall
With the recent booming of artificial intelligence (AI), particularly deep learning techniques, digital healthcare is one of the prevalent areas that could gain benefits from AI-enabled functionality. This research presents a novel AI-enabled Internet of Things (IoT) device operating from the ESP-8266 platform capable of assisting those who suffer from impairment of hearing or deafness to communicate with others in conversations. In the proposed solution, a server application is created that leverages Google’s online speech recognition service to convert the received conversations into texts, then deployed to a micro-display attached to the glasses to display the conversation contents to deaf people, to enable and assist conversation as normal with the general population. Furthermore, in order to raise alert of traffic or dangerous scenarios, an ‘urban-emergency’ classifier is developed using a deep learning model, Inception-v4, with transfer learning to detect/recognize alerting/alarming sounds, such as a horn sound or a fire alarm, with texts generated to alert the prospective user. The training of Inception-v4 was carried out on a consumer desktop PC and then implemented into the AI-based IoT application. The empirical results indicate that the developed prototype system achieves an accuracy rate of 92% for sound recognition and classification with real-time performance.
{"title":"A Deep Learning Based Wearable Healthcare Iot Device for AI-Enabled Hearing Assistance Automation","authors":"Fraser Young, Li Zhang, Richard Jiang, Han Liu, Conor Wall","doi":"10.1109/ICMLC51923.2020.9469537","DOIUrl":"https://doi.org/10.1109/ICMLC51923.2020.9469537","url":null,"abstract":"With the recent booming of artificial intelligence (AI), particularly deep learning techniques, digital healthcare is one of the prevalent areas that could gain benefits from AI-enabled functionality. This research presents a novel AI-enabled Internet of Things (IoT) device operating from the ESP-8266 platform capable of assisting those who suffer from impairment of hearing or deafness to communicate with others in conversations. In the proposed solution, a server application is created that leverages Google’s online speech recognition service to convert the received conversations into texts, then deployed to a micro-display attached to the glasses to display the conversation contents to deaf people, to enable and assist conversation as normal with the general population. Furthermore, in order to raise alert of traffic or dangerous scenarios, an ‘urban-emergency’ classifier is developed using a deep learning model, Inception-v4, with transfer learning to detect/recognize alerting/alarming sounds, such as a horn sound or a fire alarm, with texts generated to alert the prospective user. The training of Inception-v4 was carried out on a consumer desktop PC and then implemented into the AI-based IoT application. The empirical results indicate that the developed prototype system achieves an accuracy rate of 92% for sound recognition and classification with real-time performance.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126795987","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 : 2020-05-04DOI: 10.1109/ICMLC51923.2020.9469532
David Lonsdale, Li Zhang, Richard Jiang
In this paper, we present our work on developing robot arm prosthetic via deep learning. Our work proposes to use transfer learning techniques applied to the Google Inception model to retrain the final layer for surface electromyography (sEMG) classification. Data have been collected using the Thalmic Labs Myo Armband and used to generate graph images comprised of 8 subplots per image containing sEMG data captured from 40 data points per sensor, corresponding to the array of 8 sEMG sensors in the armband. Data captured were then classified into four categories (Fist, Thumbs Up, Open Hand, Rest) via using a deep learning model, Inception-v3, with transfer learning to train the model for accurate prediction of each on real-time input of new data. This trained model was then downloaded to the ARM processor based embedding system to enable the brain-controlled robot-arm prosthetic manufactured from our 3D printer. Testing of the functionality of the method, a robotic arm was produced using a 3D printer and off-the-shelf hardware to control it. SSH communication protocols are employed to execute python files hosted on an embedded Raspberry Pi with ARM processors to trigger movement on the robot arm of the predicted gesture.
{"title":"3D Printed Brain-Controlled Robot-Arm Prosthetic via Embedded Deep Learning From sEMG Sensors","authors":"David Lonsdale, Li Zhang, Richard Jiang","doi":"10.1109/ICMLC51923.2020.9469532","DOIUrl":"https://doi.org/10.1109/ICMLC51923.2020.9469532","url":null,"abstract":"In this paper, we present our work on developing robot arm prosthetic via deep learning. Our work proposes to use transfer learning techniques applied to the Google Inception model to retrain the final layer for surface electromyography (sEMG) classification. Data have been collected using the Thalmic Labs Myo Armband and used to generate graph images comprised of 8 subplots per image containing sEMG data captured from 40 data points per sensor, corresponding to the array of 8 sEMG sensors in the armband. Data captured were then classified into four categories (Fist, Thumbs Up, Open Hand, Rest) via using a deep learning model, Inception-v3, with transfer learning to train the model for accurate prediction of each on real-time input of new data. This trained model was then downloaded to the ARM processor based embedding system to enable the brain-controlled robot-arm prosthetic manufactured from our 3D printer. Testing of the functionality of the method, a robotic arm was produced using a 3D printer and off-the-shelf hardware to control it. SSH communication protocols are employed to execute python files hosted on an embedded Raspberry Pi with ARM processors to trigger movement on the robot arm of the predicted gesture.","PeriodicalId":170815,"journal":{"name":"2020 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114774557","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：