Research Of Human Identification On Continuous Cardiac Cycle ECG Signal

The traditional identification technology has been widely recognized in the field of security because of its easy to lose,copy,steal and other defects.Compared with other ECG signals such as iris,fingerprint,DNA and other traditional biological characteristics,it has a higher safety factor because of its mechanism is complex and difficult to plagiarism and imitation.Because of the unique characteristics of ECG signal,it has a broad application prospect in the field of the security of human identity recognition.In this paper,based on the analysis of the characteristics of ECG signal and the problems existing in the current research,the ECG identification method based on continuous cardiac cycle is proposed.First of all,the ECG identification method based on the gravity of the PRT triangle is proposed in this thesis.The method use the highest accuracy rate of PRT in ECG signal fiducial points,define the PRT gravity of the ECG signal.By extracting the features of the continuous gravity of the PRT triangle,the distribution boundary values of the extracted features are used as the judging conditions to human identify.This method data obtained from the ECG signals of 25 healthy college students.According to the ECG signal from 25 healthy college students,in the training set(200)and test set(196)nearly 1:1 data rate on the gravity of PRT vertex distribution discrimination experiments.The experimental results show that the average correct recognition rate(CRR)is 94.12% and the average false acceptance rate(FAR)is 4.11%.In order to solve the boundary conditions classification method is on sensitive to the waveform quality and training samples quantity,this paper proposes ECG identification method based on continuous gravity of PRT channel.The method of continuous cardiac the gravity of PRT channel is described as a matrix.By extracting the features of the continuous gravity of the triangle PRT matrix features to achieve dentity recognition.Finally,we used the 25 healthy students data from each of the first 100 consecutive ECG triangle channel as a template,followed by up to 280 consecutive ECG triangle channel as a test set,6 specific experimental methods are divided into.On the same experimental data,the average CRR of matrix operations method is 99.94%,and the recognition rate is 1.12% higher than that of the correlation coefficient.The results show that the proposed method can prove the stability and the difference of the ECG signal.