It is vital to recognize the intention of finger motions for human-machine interaction(HMI).The latest research focuses on fine myoelectric control through the decoding of neural motor unit action potential trains(MUA...It is vital to recognize the intention of finger motions for human-machine interaction(HMI).The latest research focuses on fine myoelectric control through the decoding of neural motor unit action potential trains(MUAPt) from high-density surface electromyographic(sEMG) signals.However,the existing EMG decoding algorithms rarely obtain the spatial matching relationship between decoded motion units(MU) and designated muscles,and the control interface can only recognize the trained hand gestures.In this study,a semi-supervised HMI based on MU-muscle matching(MMM) is proposed to recognize individual finger motions and even the untrained combined multi-finger actions.Through automatic channel selection from high-density s EMG signals,the optimal spatial positions to monitor the MU activation of finger muscles are determined.Finger tapping experiment is carried out on ten subjects,and the experimental results show that the proposed s EMG decomposition algorithm based on MMM can accurately identify single finger motions with an accuracy of 93.1%±1.4%,which is comparable to that of state-of-the-art pattern recognition methods.Furthermore,the MMM allows unsupervised recognizing the untrained combined multi-finger motions with an accuracy of 73%±3.8%.The outcomes of this study benefit the practical applications of HMI,such as controlling prosthetic hand and virtual keyboard.展开更多
We advance here a novel methodology for robust intelligent biometric information management with inferences and predictions made using randomness and complexity concepts. Intelligence refers to learning, adap- tation,...We advance here a novel methodology for robust intelligent biometric information management with inferences and predictions made using randomness and complexity concepts. Intelligence refers to learning, adap- tation, and functionality, and robustness refers to the ability to handle incomplete and/or corrupt adversarial information, on one side, and image and or device variability, on the other side. The proposed methodology is model-free and non-parametric. It draws support from discriminative methods using likelihood ratios to link at the conceptual level biometrics and forensics. It further links, at the modeling and implementation level, the Bayesian framework, statistical learning theory (SLT) using transduction and semi-supervised lea- rning, and Information Theory (IY) using mutual information. The key concepts supporting the proposed methodology are a) local estimation to facilitate learning and prediction using both labeled and unlabeled data;b) similarity metrics using regularity of patterns, randomness deficiency, and Kolmogorov complexity (similar to MDL) using strangeness/typicality and ranking p-values;and c) the Cover – Hart theorem on the asymptotical performance of k-nearest neighbors approaching the optimal Bayes error. Several topics on biometric inference and prediction related to 1) multi-level and multi-layer data fusion including quality and multi-modal biometrics;2) score normalization and revision theory;3) face selection and tracking;and 4) identity management, are described here using an integrated approach that includes transduction and boosting for ranking and sequential fusion/aggregation, respectively, on one side, and active learning and change/ outlier/intrusion detection realized using information gain and martingale, respectively, on the other side. The methodology proposed can be mapped to additional types of information beyond biometrics.展开更多
基金supported in part by the China National Key R&D Program(Grant No.2018YFB1307200)the National Natural Science Foundation of China (Grant Nos.51905339&91948302)。
文摘It is vital to recognize the intention of finger motions for human-machine interaction(HMI).The latest research focuses on fine myoelectric control through the decoding of neural motor unit action potential trains(MUAPt) from high-density surface electromyographic(sEMG) signals.However,the existing EMG decoding algorithms rarely obtain the spatial matching relationship between decoded motion units(MU) and designated muscles,and the control interface can only recognize the trained hand gestures.In this study,a semi-supervised HMI based on MU-muscle matching(MMM) is proposed to recognize individual finger motions and even the untrained combined multi-finger actions.Through automatic channel selection from high-density s EMG signals,the optimal spatial positions to monitor the MU activation of finger muscles are determined.Finger tapping experiment is carried out on ten subjects,and the experimental results show that the proposed s EMG decomposition algorithm based on MMM can accurately identify single finger motions with an accuracy of 93.1%±1.4%,which is comparable to that of state-of-the-art pattern recognition methods.Furthermore,the MMM allows unsupervised recognizing the untrained combined multi-finger motions with an accuracy of 73%±3.8%.The outcomes of this study benefit the practical applications of HMI,such as controlling prosthetic hand and virtual keyboard.
文摘We advance here a novel methodology for robust intelligent biometric information management with inferences and predictions made using randomness and complexity concepts. Intelligence refers to learning, adap- tation, and functionality, and robustness refers to the ability to handle incomplete and/or corrupt adversarial information, on one side, and image and or device variability, on the other side. The proposed methodology is model-free and non-parametric. It draws support from discriminative methods using likelihood ratios to link at the conceptual level biometrics and forensics. It further links, at the modeling and implementation level, the Bayesian framework, statistical learning theory (SLT) using transduction and semi-supervised lea- rning, and Information Theory (IY) using mutual information. The key concepts supporting the proposed methodology are a) local estimation to facilitate learning and prediction using both labeled and unlabeled data;b) similarity metrics using regularity of patterns, randomness deficiency, and Kolmogorov complexity (similar to MDL) using strangeness/typicality and ranking p-values;and c) the Cover – Hart theorem on the asymptotical performance of k-nearest neighbors approaching the optimal Bayes error. Several topics on biometric inference and prediction related to 1) multi-level and multi-layer data fusion including quality and multi-modal biometrics;2) score normalization and revision theory;3) face selection and tracking;and 4) identity management, are described here using an integrated approach that includes transduction and boosting for ranking and sequential fusion/aggregation, respectively, on one side, and active learning and change/ outlier/intrusion detection realized using information gain and martingale, respectively, on the other side. The methodology proposed can be mapped to additional types of information beyond biometrics.
