Vein recognition method based texture features

Abstract

The invention discloses a vein recognition method based on texture features. The method includes the steps of firstly, creating a hand vein image library, images of each category being randomly divided into a training set and a testing set, and the number of images in the training set being the same as the number of images in the testing set; secondly, acquiring and pre-processing hand vein images; and thirdly, recognizing the vein images based on the texture features. The method of the invention can take the thickness of the blood vessel into full consideration, can significantly reduce the special vector storage space, is high in processing speed, has strong robustness at the same time, and finally effectively improves the recognition rate.

Description

technical field [0001] The invention relates to a recognition method, in particular to a vein recognition method based on ridge features. Background technique [0002] With the continuous development of communication equipment and Internet technology, information has spread in every corner of people's lives. Related to it, the importance of information security has been raised to an unprecedented height. Security inspection, justice, finance, mobile network, e-commerce, human-computer interaction and many other occasions require a complete authentication system to ensure information security, and the first prerequisite for ensuring security is to know who a person is. Therefore, the application of identification technology will become more and more important. [0003] Traditional identification methods include IC cards, keys, certificates, passwords, etc. Although these methods are practical, they also have obvious disadvantages. For example, certificates are easy to be lo...

AI Technical Summary

Problems solved by technology

Intuitively speaking, the ridge feature is the most obvious feature of the human eye, but since there are generally only 3-5 main blood vessels on the back of the hand, the relationship between them is simple, and there is no complicated intersection, so the available information based on the ridge feature is compared with The degree of distinction is not high enough for the vein recognition of tens of thousands of people

Moreover, researchers are currently focusing on thinning and skeletonizing blood vessels, transforming vein recognition into a problem similar to text recognition, and ignoring the fact that the thickness of blood vessels is also a feature

Images