Identity recognition method, device and system based on electrocardiogram signals

Abstract

The invention provides an identity recognition method, device and system based on electrocardiogram signals. The method includes: receiving the electrocardiogram signals of a to-be-recognized user; performing filtering and noise reduction on the electrocardiogram signals of the to-be-recognized user; extracting the primary feature parameters of the electrocardiogram signals subjected to the filtering and noise reduction; using a principal component analysis method to screen the primary feature parameters so as to obtain final feature parameters; using the final feature parameters to construct the electrocardiogram signal feature vectors of the to-be-recognized user; comparing the electrocardiogram signal feature vectors of the to-be-recognized user with a pre-stored feature vector template; judging whether the electrocardiogram signal feature vectors of the to-be-recognized user is consistent with the pre-stored feature vector template or not, and if so, determining that the identity of the to-be-recognized user is successfully recognized. The method has the advantages that the method is especially suitable for implemented on mobile equipment such as smart phones and tablet computers which are relatively weak in computing ability, usability of using the mobile equipment to perform identity recognition based on the electrocardiogram signals is increased, and user satisfaction degree is enhanced.

Description

technical field [0001] The present invention relates to the technical field of identification, and in particular, to an identification method, device and system based on an electrocardiogram signal. Background technique [0002] In the modern society with the rapid development of network information technology, the protection of personal information security is more and more important. Identity recognition technology has been widely used as an information security technology. Traditional identification methods, such as personal certificates, keys and other authentication methods based on cryptography, are easily stolen and forged by others. Biometric Identification Technology (BIT) is a branch of pattern recognition and a new means of authentication. By combining computer with optics, acoustics, biosensors, and biometrics, it is a technology that uses the inherent biological or behavioral characteristics of the human body to identify individuals. Each individual has a uni...

AI Technical Summary

Problems solved by technology

In practice, when the ECG signal is collected and converted from analog to digital, many different interferences and noises will be mixed into it, such as the power frequency interference of the equipment itself, the myoelectric interference of the patient's body, and the human body during ECG sampling. Changes, etc.; any of them will cause ECG waveform deformation and drift, so that the next step of feature extraction can not achieve the expected effect

[0017] 2. Although patent 1 proposes filtering and denoising preprocessing of ECG signals, the proposed protection content is not very targeted

In the preprocessing step of the ECG signal, it is only explained that the filtering method based on the passband filtering method, wavelet transform filtering, Hilbert-Huang transform and adaptive morphological filtering method is to be used, without considering the complexity of the filtering algorithm and its adaptability in specific systems

[0018] 3. None of the above three patents proposed to perform quality inspection on the filtered ECG signal

The ECG signal after denoising processing may be mutated. If the eigenvalues ​​are extracted based on the mutated ECG signal, the waveform quality of the ECG signal may not be guaranteed and the accuracy of identification will be reduced.

[0019] 4. Patent No. 1 (which mentions template update) and Patent No. 3 did not give instructions on the update of the ECG template library

Although Patent No. 2 provides a template update system, it will be automatically updated every time you log in. In real life, sometimes the template is updated by using the ECG signal when the user's physical condition is not good, and after the user's physical condition recovers, Cases where identification cannot be carried out smoothly

[0020] 5. The electrocardiographic signal

Images