Dimitrios Hatzinakos

QoS and energy-aware dynamic routing in Wireless Multimedia Sensor Networks

The increasing availability of low-cost hardware along with the rapid growth of wireless devices has enabled the development of Wireless Multimedia Sensor Networks (WMSNs). Multimedia content such as video and audio streaming is transmitted over a WMSN which can easily be deployed with low cost. However, enabling real-time data applications in those networks demands not only Quality of Service (QoS) awareness, but also efficient energy management. Sensor network devices have limited energy resources. The limited energy poses significant threats on the QoS of WMSNs. In this paper, to improve the efficiency of QoS-aware routing, we examine an angle-based QoS and energy-aware dynamic routing scheme designed for WMSNs. The proposed approach uses the inclination angle and the transmission distance between nodes to optimize the selection of the forwarding candidate set and extend network lifetime. Simulation results indicate that considerable lifetime values can be achieved.

Continuous Authentication Using One-Dimensional Multi-Resolution Local Binary Patterns (1DMRLBP) in ECG Biometrics

The objective of a continuous authentication system is to continuously monitor the identity of subjects using biometric systems. In this paper, we proposed a novel feature extraction and a unique continuous authentication strategy and technique. We proposed One-Dimensional Multi-Resolution Local Binary Patterns (1DMRLBP), an online feature extraction for one-dimensional signals. We also proposed a continuous authentication system, which uses sequential sampling and 1DMRLBP feature extraction. This system adaptively updates decision thresholds and sample size during run-time. Unlike most other local binary patterns variants, 1DMRLBP accounts for observations temporal changes and has a mechanism to extract one feature vector that represents multiple observations. 1DMRLBP also accounts for quantization error, tolerates noise, and extracts local and global signal morphology. This paper examined electrocardiogram signals. When 1DMRLBP was applied on the University of Toronto database (UofTDB) 1,012 single session subjects database, an equal error rate (EER) of 7.89% was achieved in comparison to 12.30% from a state-of-the-art work. Also, an EER of 10.10% was resulted when 1DMRLBP was applied to UofTDB 82 multiple sessions database. Experiments showed that using 1DMRLBP improved EER by 15% when compared with a biometric system based on raw time-samples. Finally, when 1DMRLBP was implemented with sequential sampling to achieve a continuous authentication system, 0.39% false rejection rate and 1.57% false acceptance rate were achieved.

Differential components of discriminative 2D Gaussian–Hermite moments for recognition of facial expressions

Abstract This paper deals with a new expression recognition method by representing facial images in terms of higher-order two-dimensional orthogonal Gaussian–Hermite moments (GHMs) and their geometric invariants. Only the moments having high discrimination power are selected as a set of features for expressions. To obtain the differentially expressive components of the moments, the discriminative GHMs are projected on to a new expression-invariant subspace using the correlations among the neutral faces. Features obtained from the discriminative moments and differentially expressive components of the moments are used to recognize an expression using the well-known support vector machine classifier. Experimental results presented are obtained from commonly-referred databases such as the CK-AUC, FRGC, and MMI that have posed or spontaneous expressions as well as the GENKI database that has expressions in-the-wild. Experiments on mutually exclusive subjects reveal that the performance of expression recognition of the proposed method is significantly better than that of the existing or similar methods, which use the local or patch-based high dimensional binary patterns, directional number patterns generated from derivatives of Gaussian, Gabor- or other moment-based features.

The effect of Propofol induced anesthesia on human 40-Hz auditory steady state response

40-Hz auditory steady state responses (ASSR)s recorded from 12 human subjects during Propofol-induced anesthesia are studied in this paper. The 40-Hz ASSR signals are recorded in 8 channel stimulated electroencephalogram (EEG). The ASSR sweeps are extracted from 300 stimulated EEG epochs and updated every 0.5 seconds. Variations of the signal in time and frequency in 8 different channels are investigated both in constant times before and after anesthetic injection and relative to eyelash reflex disappearance in order to achieve a consistent result among all subjects. The latter demonstrates reduction in peak to peak amplitude 40 Hz and 80 Hz components of the signals after eyelash reflex disappearance in all 8 channels for all subjects.

On biometric systems: electrocardiogram Gaussianity and data synthesis

Electrocardiogram is a slow signal to acquire, and it is prone to noise. It can be inconvenient to collect large number of ECG heartbeats in order to train a reliable biometric system; hence, this issue might result in a small sample size phenomenon which occurs when the number of samples is much smaller than the number of observations to model. In this paper, we study ECG heartbeat Gaussianity and we generate synthesized data to increase the number of observations. Data synthesis, in this paper, is based on our hypothesis, which we support, that ECG heartbeats exhibit a multivariate normal distribution; therefore, one can generate ECG heartbeats from such distribution. This distribution is deviated from Gaussianity due to internal and external factors that change ECG morphology such as noise, diet, physical and psychological changes, and other factors, but we attempt to capture the underlying Gaussianity of the heartbeats. When this method was implemented for a biometric system and was examined on the University of Toronto database of 1012 subjects, an equal error rate (EER) of 6.71% was achieved in comparison to 9.35% to the same system but without data synthesis. Dimensionality reduction is widely examined in the problem of small sample size; however, our results suggest that using the proposed data synthesis outperformed several dimensionality reduction techniques by at least 3.21% in EER. With small sample size, classifier instability becomes a bigger issue and we used a parallel classifier scheme to reduce it. Each classifier in the parallel classifier is trained with the same genuine dataset but different imposter datasets. The parallel classifier has reduced predictors’ true acceptance rate instability from 6.52% standard deviation to 1.94% standard deviation.

Enhanced binary patterns for electrocardiogram (ECG) biometrics

Most, if not all, binary patterns variants consider signals observations separately; hence, binary patterns variants ignore any relationship among observations. In this paper we proposed an algorithm that enhances binary patterns extraction to accommodate for temporal progression changes. The enhanced binary patterns feature extraction extracts a single feature vector that captures changes occurred to observations over time. This enhancement is crucial in cases where the examined signal is repetitive in nature, such as ECG signal. Enhanced binary patterns were examined for ECG biometric application on ECG database with 1,012 subjects. The enhanced binary patterns achieved an EER of 7.89% in comparison to 12.4% and 12.3% for non-enhanced binary patterns and a state of the art work. We also showed that enhanced binary patterns features are capable to extract discriminative ECG features that reduced EER by 15% when compared to time-domain raw samples.

Feature selection from multisession electrocardiogram signals for identity verification

This paper proposes a framework for human recognition based on Electrocardiogram (ECG) signals. We particularly consider a verification scenario in which only one recording session is available for enrolling a subject. Capturing the non-stationarity of ECG and constructing a robust model which can be well generalized to unseen data may not be possible via having only one training session. Under this scenario, we propose to use an auxiliary multisession ECG data set to extract a prior knowledge about the behaviour of ECG signal across sessions. A pool of different types of features is formed and a subset of good features is selected using auxiliary data set. By considering only the selected features for enrollment and test, significant performance improvement is achieved. Existing feature selection approaches are designed to be used in conventional classification problems which are based on a set of training samples and a vector of class labels. Our work is different from the previous works in that we not only consider the class labels but also consider session labels. Features selected from a multisession auxiliary data set are used as a prior knowledge to build robust templates in the enrollment stage where only one training session is available. Experimental results demonstrate effectiveness of the proposed method to cope with non-stationarity of ECG signals across different sessions.

Reference empirical mode decomposition

Proposed in this paper is a modified version of empirical mode decomposition (EMD) that guarantees unified signal representation after decomposition. Reference EMD (R-EMD), decomposes each signal with a set of reference sinusoids to achieve a wavelet-like frequency separation, while retaining the adaptive feature of the EMD algorithm. A brief proof is also provided on the role of the reference sinusoids in extracting frequencies at each level. R-EMD provides a solution for the problem of high dimensionality and complexity in decomposing multiple signals together.

40-Hz ASSR depth of anaesthesia index

A novel method for defining an index based on multi-level clustering of 40-Hz auditory steady state response is presented in this paper. The index is a measure of depth of anaesthesia which can help monitoring depth of anaesthesia more closely and accurately. Multi-level expectation maximization (EM) is used for clustering the recorded 40-Hz auditory steady state response signals recorded from human subjects. The clustering information is used to define the depth of anaesthesia index. Rather than extracting the maximum amplitude and frequency at each cycle as clustering features, principal components analysis (PCA) is used for analyzing all samples of the cycles and projecting data into a lower dimension space. Both dimension reduction and clustering schemes are unsupervised methods, hence the algorithm does not need initial data labeling or training phase.

Mutual information-based selection of audiovisual affective features to predict instantaneous emotional state

Automatic prediction of continuous level emotional state requires selection of suitable affective features to develop a regression system based on supervised machine learning. This paper investigates the performance of low-level dynamic features for predicting two common dimensions of emotional state, namely, valence and arousal instantaneously. Low-complexity features are extracted from audio and visual modalities independently and fused in the feature level. Features with minimum redundancy and maximum relevancy are chosen by using the mutual information-based selection process. The performance of frame-by-frame prediction of emotional state using the moderate length features as proposed in this paper is evaluated on spontaneous and naturalistic human-human conversation of SEMAINE database. Experimental results show that the proposed features selected by mutual information can be used for instantaneous prediction of emotional state with an accuracy higher than traditional audio or visual features that are used for affective computation.