Person detection by using ECG signals

Recently, numerous researches have been executed to create reliable systems to recognize persons based on their biometric information. Person Identification (PI) systems have become popular among researchers using different methods. Considering the unique advantages of electrocardiogram (ECG) signals, it has been an important research area that has attracted the attention of researchers for person identification studies. Since ECG data are difficult to imitate or copy, these data become more advantageous than other data in biometric studies such as person identification. In this study, the One-Dimensional Multilayer Co-occurrence Matrices (1D-MLGLCM) method was developed, which enables to obtain a strong feature set from one-dimensional signals and this method has been applied to recognize individuals based on ECG signals. The dataset used in the experiments contains ECG data of 90 subjects whose ages ranged from 13 to 75 years. In the study, first of all, ECG signals are normalized at 32 different intervals for person identification system. One-Dimensional Co-occurrence Matrices (1D-GLCM) method is applied to each of these multiplex signals obtained to obtain co-occurrence matrices. Then, the Haralick features of these matrices are obtained to be given as inputs to classification algorithms such as Random Forest (RF), Support Vector Machine (SVM), Naive Bayes (NB), Bayes Net (BN) and K-Nearest Neighbors (KNN). The proposed method achieved a success rate of 93,414% using SVM. In conclusion, the study proves that the proposed method achieves very effective results in person identification problems by using ECG signals.