A model-based approach to human identification using ECG

Homer, Mark; Irvine, John; Wendelken, Suzanne M.

doi:10.1117/12.819327

Cited by 11 publications

(10 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most of the methods based on non-fiducial features require the detection of the R peaks for heartbeat segmentation and alignment [40], [37], [12], [39], [14], [43], [55], [48], [27], [57], [51], [33], [50], [53], [32], [18], [11].…”

Section: Survey Of Ecg Recognition Methodsmentioning

confidence: 99%

“…(NNC) as a special case (k = 1), is the most frequently used classifier type in the ECG recognition literature [16], [26], [15], [17], [23], [25], [51], [54], [56], [13]. It involves comparing a feature vector to a collection of feature vectors, and selecting the top k vectors that produce the best match.…”

Section: ) Knn Classifiers: K Nearest Neighbors (Knn) Classifiers Wmentioning

confidence: 99%

“…The database used by Irvine et al [7], [34], [43], [24], Israel et al [21], Homer et al [51], and Jang et al [55] were from data collection protocols designed to elicit different states of anxiety in the individuals.…”

Section: B Private Databasesmentioning

confidence: 99%

See 2 more Smart Citations

ECG Biometric Recognition: A Comparative Analysis

Odinaka

Lai

Kaplan

et al. 2012

IEEE Trans.Inform.Forensic Secur.

300

193

View full text Add to dashboard Cite

The electrocardiogram (ECG) is an emerging biometric modality that has seen about thirteen years of development, in peer-reviewed literature, and as such deserves a systematic review and discussion of the associated methods and findings. In this paper, we review most of the techniques that have been applied to the use of the electrocardiogram for biometric recognition. In particular, we categorize the methodologies based on the features and the classification schemes. Finally, a comparative analysis of the authentication performance of a few of the ECG biometric systems is presented, using our in-house database. The comparative study includes the cases where training and testing data come from the same and different sessions (days). The authentication results show that most of the algorithms that have been proposed for ECG based biometrics perform well, when the training and testing data come from the same session. However, when training and testing data come from different sessions, a performance degradation occurs. Multiple training sessions were incorporated to diminish the loss in performance. That notwithstanding, only a few of the proposed ECG recognition algorithms appear to be able to support performance improvement due to multiple training sessions. Only three of these algorithms produced This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. 2 equal error rates (EERs) in the single digits, including an EER of 5.5% using a method proposed by us.

show abstract

Section: Survey Of Ecg Recognition Methodsmentioning

confidence: 99%

Section: ) Knn Classifiers: K Nearest Neighbors (Knn) Classifiers Wmentioning

confidence: 99%

See 1 more Smart Citation

ECG Biometric Recognition: A Comparative Analysis

Odinaka

Lai

Kaplan

et al. 2012

IEEE Trans.Inform.Forensic Secur.

300

193

View full text Add to dashboard Cite

show abstract

“…For example in [1] only STFT features is used. Moreover, our data set consists 81 subjects and The average number of sessions recorded for each subject is 4.28 with standard deviation of 0.97 which is relatively bigger than many of previously reported data sets with multiple sessions [2], [3], [4], [5], [6], [7], [8].…”

Section: Introductionmentioning

confidence: 97%

Feature selection from multisession electrocardiogram signals for identity verification

Komeili

Armanfard

Hatzinakos

et al. 2015

2015 IEEE 28th Canadian Conference on Electrical and Computer Engineering (CCECE)

View full text Add to dashboard Cite

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.

show abstract

“…Unlike other Biometric technologies, ECG has the inherent property of life detection for identification [6]. In the recent past, many studies have been used for ECG based biometric recognition systems [7][8][9][10][11]. In the literature, different techniques have been used for ECG biometrics.…”

Section: Introductionmentioning

confidence: 99%

Untitled

2017

JESTR

View full text Add to dashboard Cite

Person identification is an essential task in defense and forensic applications such as surveillance and criminal investigation activities. Studies conducted in the past included different types of regular biometric traits, namely fingerprint, face, iris, and voice have a limitation of falsification, and they do not fit guarantee of liveness of the subject. In this context, Electrocardiogram based Biometric recognition is an alternative solution, where the security of the information is very high level. This research aims to provide with a complete systematic approach to ECG based Biometric recognition, which contains primarily the processing of raw signal through noise elimination filters and a time domain analysis is carried for all ECG characteristic waves detection. Subsequently, an effective feature extraction method for ECG is developed, which extracts six best P-QRS-T fragments based on priority and their positions are normalized. Also, 72-time domain features are calculated. These features are formed into feature vector corresponding to each signal separately for both train and test data sets. To analyze the performance of the system, the feature vectors are trained with various Machine learning classification algorithms like Artificial Neural Network (ANN), Support Vector Machine (SVM) and K-Nearest Neighbor (KNN). Finally, the proposed system is tested with a challenging, public available MIT-BIH ECG-ID Database. A comparative analysis using performance parameters is made with different classifiers, and the obtained results show that SVM classifier provides 93.709% overall classification accuracy when compared to previous literature.

show abstract

A model-based approach to human identification using ECG

Cited by 11 publications

References 0 publications

ECG Biometric Recognition: A Comparative Analysis

ECG Biometric Recognition: A Comparative Analysis

Feature selection from multisession electrocardiogram signals for identity verification

Untitled

Contact Info

Product

Resources

About