Biometric characteristic identification method and device

Abstract

The invention discloses a biometric characteristic identification method. The biometric characteristic identification method includes the following steps that: a biometric characteristic image to be identified is received; a biometric characteristic identification model is adopted to identify a biometric characteristic to be identified, and whether the biometric characteristic to be identified is matched with a preset biometric characteristic image is determined, wherein the biometric characteristic identification model is obtained through training a convolutional neural network by adopting a biometric characteristic training sample; when the biometric characteristic to be identified is matched with the preset biometric characteristic image, the biometric characteristic image to be identified is stored; and when the stored biometric characteristic image to be identified is larger than a preset threshold value, and the biometric characteristic identification model is updated. The invention also provides a biometric characteristic identification device. With the biometric characteristic identification device adopted, high precision of the extraction of biometric characteristics from a biometric characteristic image can be ensured, and a biometric characteristic identification result can be more accurate.

Description

technical field [0001] The invention belongs to the field of biological feature identification, and more specifically relates to a biological feature identification method and device. Background technique [0002] With the introduction of biometric identification technology into smart mobile devices by Apple, biometric identification has become the mainstream configuration of smart terminals. Biometric identification technology includes: two stages of registration and verification. In the registration phase, the biometric information of the user is collected, and the characteristic value in the biometric information is extracted, and stored in the memory for later biometric verification; in the verification phase, the biometric information of the user to be verified is collected, and the biometric information is extracted The eigenvalues ​​in , and then match with the preset eigenvalues. If the match is successful, the next operation can be performed; otherwise, the operat...

AI Technical Summary

Problems solved by technology

[0003] When the fingers are stained with oil, sweat or peeling, the collected biometric information is often different from the preset biometric information, which leads to a greatly reduced biometric recognition rate.

Images