Search

Emotion recognition using EEG has been widely studied to address the challenges associated with affective computing. Using manual feature extraction methods on EEG signals results in sub-optimal performance by the learning models. With the advancements in deep learning as a tool for automated feature engineering, in this work, a hybrid of manual and automatic feature extraction methods has been proposed. The asymmetry in different brain regions is captured in a 2D vector, termed the AsMap, from the differential entropy features of EEG signals. These AsMaps are then used to extract features automatically using a convolutional neural network model. The proposed feature extraction method has been compared with differential entropy and other feature extraction methods such as relative asymmetry, differential asymmetry and differential caudality. Experiments are conducted using the SJTU emotion EEG dataset and the DEAP dataset on different classification problems based on the number of classes. Results obtained indicate that the proposed method of feature extraction results in higher classification accuracy, outperforming the other feature extraction methods. The highest classification accuracy of 97.10% is achieved on a three-class classification problem using the SJTU emotion EEG dataset. Further, this work has also assessed the impact of window size on classification accuracy., Comment: 18 pages, 7 figures, published in Sensors (MDPI)

Electroencephalogram (EEG) signals may get easily contaminated by muscle artifacts, which may lead to wrong interpretation in the brain--computer interface (BCI) system as well as in various medical diagnoses. The main objective of this paper is to remove muscle artifacts without distorting the information contained in the EEG. A novel multi-stage EEG denoising method is proposed for the first time in which wavelet packet decomposition (WPD) is combined with a modified non-local means (NLM) algorithm. At first, the artifact EEG signal is identified through a pre-trained classifier. Next, the identified EEG signal is decomposed into wavelet coefficients and corrected through a modified NLM filter. Finally, the artifact-free EEG is reconstructed from corrected wavelet coefficients through inverse WPD. To optimize the filter parameters, two meta-heuristic algorithms are used in this paper for the first time. The proposed system is first validated on simulated EEG data and then tested on real EEG data. The proposed approach achieved average mutual information (MI) as 2.9684 $\pm$ 0.7045 on real EEG data. The result reveals that the proposed system outperforms recently developed denoising techniques with higher average MI, which indicates that the proposed approach is better in terms of quality of reconstruction and is fully automatic., Comment: 21 pages, 9 figures, 9 tables

Efficiently implementing remote sensing image classification with high spatial resolution imagery can provide a significant value in Land Use and Land Cover (LULC) classification. The new advances in remote sensing and deep learning technologies have facilitated the extraction of spatiotemporal information for LULC classification. Moreover, the diverse disciplines of science, including remote sensing, have utilised tremendous improvements in image classification by Convolutional Neural Networks (CNNs) with transfer learning. In this study, instead of training CNNs from scratch, the transfer learning is applied to fine-tune pre-trained networks Visual Geometry Group (VGG16) and Wide Residual Networks (WRNs), by replacing the final layer with additional layers, for LULC classification using the red-green-blue version of the EuroSAT dataset. Moreover, the performance and computational time are compared and optimised with techniques, such as early stopping, gradient clipping, adaptive learning rates, and data augmentation. The proposed approaches have addressed the limited-data problem, and very good accuracies are achieved. The results show that the proposed method based on the WRNs performs better than the previous best-stated results in terms of the computational efficiency and accuracy from 98.57% to 99.17%.

Craigen introduced and studied signed group Hadamard matrices extensively and eventually provided an asymptotic existence result for Hadamard matrices. Following his lead, Ghaderpour introduced signed group orthogonal designs and showed an asymptotic existence result for orthogonal designs and consequently Hadamard matrices. In this paper, we construct some interesting families of orthogonal designs using signed group orthogonal designs to show the capability of signed group orthogonal designs in generation of different types of orthogonal designs., Comment: To appear in Discrete Mathematics (Elsevier). No figures

Orthogonal designs and weighing matrices have many applications in areas such as coding theory, cryptography, wireless networking and communication. In this paper, we first show that if positive integer $k$ cannot be written as the sum of three integer squares, then there does not exist any skew-symmetric weighing matrix of order $4n$ and weight $k$, where $n$ is an odd positive integer. Then we show that for any square $k$, there is an integer $N(k)$ such that for each $n\ge N(k)$, there is a symmetric weighing matrix of order $n$ and weight $k$. Moreover, we improve some of the asymptotic existence results for weighing matrices obtained by Eades, Geramita and Seberry., Comment: To appear in International Journal of Combinatorics (Hindawi). in Int. J. Combin. (Feb 2016)

Hadamard matrices, orthogonal designs and amicable orthogonal designs have a number of applications in coding theory, cryptography, wireless network communication and so on. Product designs were introduced by Robinson in order to construct orthogonal designs especially full orthogonal designs (no zero entries) with maximum number of variables for some orders. He constructed product designs of orders $4$, $8$ and $12$ and types $\big(1_{(3)}; 1_{(3)}; 1\big),$ $\big(1_{(3)}; 1_{(3)}; 5\big)$ and $\big(1_{(3)}; 1_{(3)}; 9\big)$, respectively. In this paper, we first show that there does not exist any product design of order $n\neq 4$, $8$, $12$ and type $\big(1_{(3)}; 1_{(3)}; n-3\big),$ where the notation $u_{(k)}$ is used to show that $u$ repeats $k$ times. Then, following the Holzmann and Kharaghani's methods, we construct some classes of disjoint and some classes of full amicable orthogonal designs, and we obtain an infinite class of full amicable orthogonal designs. Moreover, a full amicable orthogonal design of order $2^9$ and type $\big(2^6_{(8)}; 2^6_{(8)}\big)$ is constructed., Comment: 12 pages, To appear in the Australasian Journal of Combinatorics

Craigen introduced and studied {\it signed group Hadamard matrices} extensively in \cite{Craigenthesis, Craigen}. Livinskyi \cite{Ivan}, following Craigen's lead, studied and provided a better estimate for the asymptotic existence of signed group Hadamard matrices and consequently improved the asymptotic existence of Hadamard matrices. In this paper, we introduce and study signed group orthogonal designs. The main results include a method for finding signed group orthogonal designs for any $k$-tuple of positive integer and then an application to obtain orthogonal designs from signed group orthogonal designs, namely, for any $k$-tuple $\big(u_1, u_2, ..., u_{k}\big)$ of positive integers, we show that there is an integer $N=N(u_1, u_2, ..., u_k)$ such that for each $n\ge N$, a full orthogonal design (no zero entries) of type $\big(2^nu_1,2^nu_2,...,2^nu_{k}\big)$ exists . This is an alternative approach to the results obtained in \cite{EK}., Comment: 16 pages, To appear in Algebraic Design Theory and Hadamard Matrices (ADTHM), Springer Proceeding in Mathematics and Statistics. Editor: Charles Colbourn. Springer Proceeding in Mathematics and Statistics (PROMS), 2015

In this paper, we introduce some known map projections from a model of the Earth to a flat sheet of paper or map and derive the plotting equations for these projections. The first fundamental form and the Gaussian fundamental quantities are defined and applied to obtain the plotting equations and distortions in length, shape and size for some of these map projections., Comment: 18 pages, 13 figures, notebook

We show that if G is any nilpotent, finite group, and the commutator subgroup of G is cyclic, then every connected Cayley graph on G has a hamiltonian cycle., Comment: 16 pages, 6 figures

Suppose G is a finite group of order 30p, where p is prime. We show that if S is any generating set of G, then there is a hamiltonian cycle in the corresponding Cayley graph Cay(G;S)., Comment: 23 pages, plus 18-page appendix of notes to aid the referee (or other readers)

Suppose G is a finite group, such that |G| = 27p, where p is prime. We show that if S is any generating set of G, then there is a hamiltonian cycle in the corresponding Cayley graph Cay(G;S)., Comment: 13 pages, no figures; minor revisions, including suggestions from a referee; this version is to appear in the International Journal of Combinatorics