Electrocardiogram identity recognition method

Abstract

The invention discloses an electrocardiogram identity recognition method. The method comprises the following steps: acquiring an ECG (Electrocardiogram) signal of a human body; preprocessing the collected ECG signal to obtain a clean short-period ECG signal; generalized S transformation is carried out on the short-period ECG signals, and a phase domain feature vector Y1, a time domain feature vector Y2 and a frequency domain feature vector Y3 are extracted; inputting the phase domain feature vector Y1, the time domain feature vector Y2 and the frequency domain feature vector Y3 into a sparsity-constrained nonlinear approximation model to obtain an optimal dictionary and a corresponding optimal sparse coefficient matrix, and performing lightweight processing on the optimal sparse coefficient matrix to obtain a sparse coefficient vector; and inputting the sparse coefficient vector into a trained deep neural network model based on a bidirectional long-short-term memory network for identity recognition. According to the invention, the multi-modal feature vector is extracted from the original ECG signal to serve as the input vector of the deep neural network, so that the recognition precision is improved, and the recognition rate is increased.

Description

technical field [0001] The invention relates to the technical field of identification, in particular to an identification method based on electrocardiographic signals. Background technique [0002] Network information security is of great significance to individuals, enterprises, and the country. It involves security and privacy protection in all aspects of property, reputation, and personal life, ranging from national security fields such as criminal investigation and law enforcement applications, to unlocking smartphones. , payment and other areas of life. Traditional identification methods such as passwords and ID cards are prone to theft and forgery risks, and cannot meet the requirements of high security and high privacy in financial systems, security monitoring and other fields. The application of identification technology based on modern biometric features has attracted more and more attention, such as fingerprint recognition, face recognition, voice recognition, etc...

AI Technical Summary

Problems solved by technology

When applied to the classification task of endogenous physiological signals such as ECG identification, the differences between classes are almost invisible to the naked eye, and the differences within classes caused by noise interference, individual activity states, pathological states, and psychological emotions are also invisible. However, this has a great impact on the recognition accuracy

Therefore, even classifiers with excellent performance such as support vector machines and convolutional neural networks are difficult to achieve accurate classification goals.

Secondly, too much local feature overlap increases the computational burden, and finer inter-class and intra-class differences mean longer feature learning and training time, which cannot be applied to mobile phones for real-time and fast signal processing and feedback

Images