Fingerprint direction field calculation method based on improved partial differential model

Abstract

The invention discloses a fingerprint direction field calculation method based on an improved partial differential model, and belongs to the technical field of automatic fingerprint recognition. The fingerprint direction field calculation method mainly comprises two steps: I, solving an initial direction field of fingerprint by using a basic gradient algorithm; II, smoothing the initial direction field by using the improved partial differential model. The fingerprint direction field calculation method mainly aims to improve holding items of a conventional partial differential model, that is, firstly, the situation that the initial direction field in the conventional model is taken as holding standard is improved into the situation that the direction field of previous iteration is taken as the holding standard, and secondly, a static capture singular point function in the conventional model is improved to be a dynamic capture singular point function. By adopting the fingerprint direction field calculation method disclosed by the invention, noise areas and singular point areas can be effectively distinguished, the algorithm is relatively reasonable, the iteration times can be effectively reduced, the algorithm efficiency can be improved, and the direction field can be relatively accurately calculated.

Description

technical field [0001] The invention relates to a fingerprint direction field calculation method based on an improved partial differential model, belonging to the technical fields of automatic fingerprint identification and digital image processing. Background technique [0002] Biometric identification technology is a technology for identification based on each person's unique biological and behavioral characteristics that can be sampled and measured. Because biometric features are not as easy to steal as various certificates, nor are they as easy to forget or crack as passwords and passwords, so they have unique advantages in identification and have been widely studied internationally in recent years. Automated Fingerprint Identification System (AFIS) is more and more popular due to its small size, low cost, easy operation and high reliability, and has become one of the most important biometric technologies. [0003] A general automatic fingerprint recognition system incl...

AI Technical Summary

Problems solved by technology

The gradient-based method is the most extensive and basic method for the fingerprint direction field. It is mainly divided into two parts: the calculation of the point gradient vector and the block gradient vector; however, the simple use of the gradient method cannot effectively anti-noise, especially In the large noise area, the calculation of the direction field based on the gradient is very difficult, and the calculated direction field is very inaccurate

The filtering-based method has good anti-noise performance, but there are only a limited number of filters, and the calculated direction field is not accurate enough, and this type of method needs to calculate the output of each filter and perform In comparison, the calculation is very heavy and the efficiency is low

The model-based method summarizes the rules and limitations of the fingerprint direction field from a global perspective, and uses this to estimate the direction field of the fingerprint. The calculation of the fingerprint direction field based on a polynomial model is the classic; this type of method has a strong ability to resist noise. , but it depends on accurately extracting the singular points of the fingerprint, and at the same time, it is often not well estimated in the singular point area. In addition, most models often require a large number of parameters in the calculation process

[0005] The calculation of fingerprint direction field is an important and urgent fingerprint recognition problem. The main difficulty is that it is difficult to distinguish the singular points and noise of the fingerprint, especially when the noise appears in the singular point area. The existing estimation methods all assume the basic The direction field calculated by the gradient method at the singular point is basically correct

However, the initial direction field obtained by the gradient algorithm of the fingerprint in the noise area is actually unreliable, and for the function of capturing singular points, the original algorithm has certain defects, and cannot effectively distinguish the singular point area from the noise area

Images