Intelligent vein authentication method and system having autonomous learning capability

Abstract

The invention discloses an intelligent vein authentication method and system having autonomous learning capability. The method comprises steps that model training of the full convolutional neural network depth learning model in repeated and regular global training and irregular local training modes is carried out, before online operation of the full convolutional neural network depth learning model, one global training is carried out through utilizing accumulated samples, and optimized characteristic extraction, characteristic identification and characteristic classification parameters are stored; after online operation of the model, newly-registered samples are utilized to carry out irregular local training of the model, and characteristic classification parameter optimization is completed; under the specific condition, the newly-accumulated samples are utilized to carry out regular global training of the model, and the characteristic extraction, characteristic identification and characteristic classification parameters are optimized for another time, the model is guaranteed to be the intelligent authentication system having the autonomous learning capability, and vein authentication precision is further enhanced.

Description

technical field [0001] The invention relates to the technical field of image recognition and processing, in particular to an intelligent vein authentication method and system with self-learning ability Background technique [0002] In recent years, with the rapid development of biometric technology, fingerprint authentication has been widely used. At the same time, more and more authentication methods such as iris, voice, face, palm vein, and finger vein with higher security characteristics are also used. Among them, the finger vein authentication method has been more widely used due to its advantages of high confidentiality, non-replication, convenient authentication, and small size of authentication equipment. In the prior art, finger vein authentication generally adopts various image segmentation algorithms to separate the vein image from the collected finger image to form a binary image of vein veins, and then uses a feature extraction algorithm to extract individual ve...

AI Technical Summary

Problems solved by technology

However, in practical applications, the samples to be identified and classified are dynamically increased. Only the parameters generated during the supervised learning process before the system goes online cannot accurately identify all the newly added samples to be identified, which affects the recognition accuracy of new samples.

(2) High requirements on the amount of learning data and long learning time

The deep learning model is relatively complex, and parameter optimization requires a large amount of training data for support, and multiple iterations are required to complete the training

However, the current finger vein sample data has neither a standard training data set nor a sufficient amount of training data.

(3) The accuracy of certification has not yet reached the level of precision

Images