Heart sound as a biometric

Heartbeat Signal from Facial Video for Biometric Recognition

“…Though touch based ECG [12] and PCG [14] biometrics obtained more than 90% accuracy on some local databases, we think their direct comparison against our system is biased towards their favor, as they use obtrusive touch-based sensors, which provide precise measurement of heartbeat signals, while we, using our touch-free sensor (webcam), get only estimations of those heartbeat signals that are obtained by touch based sensors. This means that it makes sense if our touch-free system, at least in this initial step of its development, does not outperform those touch-based systems.…”

“…They use the z-chirp transformation (CZT) for feature extraction and Euclidian distance matching for identification. Puha et al [14] proposed another system by analyzing cepstral coefficients in the frequency domain for feature extraction and employing a Gaussian Mixture Model (GMM) for identification. Subsequently, different methods were proposed, such as a wavelet based method in [15] and marginal spectral analysis based method in [16].…”

Heartbeat biometrics: a sensing system perspective

“…Modern microphones and recording techniques have improved sufficiently that digital heartbeat data can even support training medical students (Barrett et al, 2004). Recently, heart sounds have been exploited to classify medical maladies and as a biometric (Nigam and Priemer, 2004;Beritelli and Serrano, 2007;Phua et al, 2008). Scanlon (2001) mapped the body sound changes as a function of macro movements.…”

“…Although some researchers have proposed methods that do not require heartbeat alignment (e.g. Phua et al, 2008), most approaches include this step. The reason for alignment is that most techniques rely on features derived from the morphology, amplitude, and timing of the heartbeat, thus requiring segmentation of individual beats.…”

Desenvolvimento de um dispositivo capaz de registrar e analisar potenciais evocados auditivos nos domínios do tempo e das freqüências

“…Alguns estudos demonstraram também a eficiência desse periférico na aquisição de sinais de eletromiografia (EMG), eletrocardiografia (ECG) e na digitalização de informações eletromecânicas, para serem monitoradas em tempo real (GULER, HARDALAC e KAYMAZ, 2002;ELKER, 2004;PHUA et al, 2008).…”