Identity recognition method based on photoplethysmography

Abstract

The invention discloses an identity recognition method based on photoplethysmography. The method comprises the following steps: 1) obtaining training group data and test group data; 2) segmenting allthe photoplethysmography signal data in the training group data and the test group data into a plurality of segments containing a plurality of photoplethysmography segments; 3) converting all the segmented segments into a time-frequency characteristic energy diagram form by using continuous wavelet transform; 4) building a neural network model combining the CNN and the LSTM; 5) sending the training group picture data into a CNN and LSTM combined neural network model for training; and 6) using the trained model to classify the test group picture data, a classification result being an identity identification result, and finally, evaluating the trained CNN and LSTM combined neural network model. The method is high in safety coefficient in application, does not need manual feature extraction,is high and stable in recognition effect precision, and can be used for personal identity recognition in the fields of government institutions, financial institutions and the like.

Description

technical field [0001] The invention relates to the interdisciplinary field of information processing and computer science, in particular to an identification method based on photoplethysmography, which can be used as a means for maintaining personal information security in government agencies, financial institutions and other fields. Background technique [0002] With the development of today's society, security issues have become particularly prominent. People lose their reliability and practicability by memorizing complex passwords or traditional identification methods with electronic ciphers. This situation makes people's demand for biometric technology getting bigger. For example, the company's confidential systems, financial transactions, computer networks and access systems to security areas are still identified and authorized through ID cards or passwords. Such systems are not sufficiently secure because ID or password information is easily stolen or forgotten. The...

AI Technical Summary

Problems solved by technology

Most of the existing identification methods based on photoplethysmography are based on artificially extracted features, which will cause problems due to the differences in photoplethysmography signals between different human individuals, resulting in the recognition accuracy and generalization of the model. Reduced ability

Images