Finger vein image quality evaluation method based on network learning

Abstract

Aiming at the problem that the performance of a finger vein recognition system is greatly influenced by the quality of an acquired image, the invention provides a finger vein image quality evaluationmethod based on network learning by comprehensively considering the characteristics of a finger vein image. The method comprises the following steps: firstly, designing seven evaluation criteria of brightness uniformity, definition, area, position offset, information entropy, contrast ratio and equivalent number of views for an acquired finger vein image to carry out image quality evaluation, andobtaining seven corresponding quality evaluation scores; normalizing the seven quality evaluation scores so as to avoid overlarge order-of-magnitude difference; and finally, taking the normalized image quality evaluation score as network input, and designing an MEA-BP-Adaboost strong classifier to obtain a vein image total quality evaluation grade. According to the method, a new solution is provided for the problem that the finger vein image quality greatly influences the recognition precision, the quality of the to-be-recognized image is evaluated according to the image quality evaluation index, the consistency of the finger vein images collected in different environments can be improved, and therefore the subsequent matching recognition accuracy of a vein recognition system is improved.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition, in particular to a method for evaluating the quality of finger vein images based on network learning. Background technique [0002] The finger vein recognition technology uses the imaging of the internal blood vessels of the finger, which is not affected by the surface conditions of the finger. Because the imaging uses near-infrared light to transmit blood vessels, the hemoglobin in the blood vessels absorbs infrared rays to form blood vessel lines. Once the individual is inactivated, the vein lines cannot be collected. With the characteristics of living body collection, the anti-counterfeiting and security are extremely high. [0003] However, during the collection process, due to lighting, improper placement of fingers, and sensor noise, etc., the vein image is degraded or even unrecognizable by the subsequent authentication system, which seriously affects the performance of the vei...

AI Technical Summary

Problems solved by technology

[0003] However, during the collection process, due to lighting, improper placement of fingers, and sensor noise, etc., the vein image is degraded or even unrecognizable by the subsequent authentication system, which seriously affects the performance of the vein authentication system.

Images