Motor imagery classification is a crucial task in the field of brain-computer interfaces (BCI)which can benefit individuals with disabilities in their movements of their limbs. However, accurately classifying such sig...
详细信息
Many existing works represent signals by covariance matrices and then develop learning methods on the Riemannian symmetric positive-definite (SPD) manifold to deal with such data. However, they summa-rize each instanc...
详细信息
Many existing works represent signals by covariance matrices and then develop learning methods on the Riemannian symmetric positive-definite (SPD) manifold to deal with such data. However, they summa-rize each instance with a single covariance matrix, omitting some potential important information, such as the time evolution of the correlation in signals. In this paper, we represent each instance by a sequence of covariance matrices and develop a novel dynamic generalized learning Riemannian space quantization (DGLRSQ) method to deal with such data representations. The proposed DGLRSQ method incorporates short-term memory mechanism in generalized learning Riemannian space quantization (GLRSQ), which is an extension of Euclidean generalized learningvectorquantization to deal with SPD matrix-valued data. The proposed method can capture the temporal evolution of the correlation in signals and thus provides better performance to its the counterpart - GLRSQ, which treats each instance as a signal covariance matrix. Empirical investigations on synthetic data and motor imagery EEG data show the superior perfor-mance of the proposed method. (c) 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://***/licenses/by-nc-nd/4.0/ )
The measurement of the physical and chemical ("physicochemical") properties of nanomaterials used in industry and science including chemistry, pharmacy, medicine, toxicology, etc., is time-consuming, expensi...
详细信息
The measurement of the physical and chemical ("physicochemical") properties of nanomaterials used in industry and science including chemistry, pharmacy, medicine, toxicology, etc., is time-consuming, expensive and requires a lot of experience of a well trained lab staff. Near-infrared spectroscopy (NIR;4.000-12.000 cm(-1)), working in the wavelength region with the highest IR energy, allows obtaining multifactorial information of the material under investigation due to the occurrence of a high number of combination and overtone vibrations. Coupling of an optimized and well-designed measurement technique with multivariate data analysis (MVA) leads to a non-destructive, fast, reliable and robust novel NIR technique for the fast and non-invasive physicochemical characterization, which is suitable for high-throughput quality control due to the short analyses times of only a few seconds. In the following chapters, the patented basic NIR techniques full-filling these aims are introduced, described, summarized and critically discussed.
暂无评论