Finger vein feature extraction method

Abstract

The invention discloses a finger vein feature extraction method. According to the method, the stability of vein feature extraction can be improved, more features can be obtained, and therefore the vein recognition accuracy can be improved conveniently. The method includes the steps that under irradiation of an infrared light source, an infrared camera is used for acquiring finger vein images; the finger vein images with brightness meeting requirements are obtained by adjusting parameters of the infrared light source and parameters of the infrared camera; a plurality of images to be processed are acquired by the adoption of the determined parameters of the light source and the determined parameters of the camera and converted into binary images of vein lines; vein existing points are sought for by combining the binary images of the vein lines, and a synthetic vein image is formed; finally, image features of finger veins are extracted from the synthetic vein image.

Description

technical field [0001] The invention relates to a finger vein feature extraction method, which belongs to the technical field of biometric feature recognition. Background technique [0002] Finger vein recognition technology is a new type of biometric identification technology. Because the user's finger vein characteristics are difficult to be forged, it has a high level of security and is especially suitable for use in places with high security requirements. [0003] The process of finger vein recognition is that the infrared light generated by the infrared light source is irradiated on the finger of the collector, filtered by the filter and reflected by the mirror, and collected by the infrared camera. Vein features are extracted from the collected images, and then feature matching is performed, thus realizing finger vein recognition. [0004] One of the key steps in finger vein recognition is vein feature extraction. The accuracy of vein feature extraction is determined...

AI Technical Summary

Problems solved by technology

However, there is still no better brightness adjustment scheme

[0005] Due to a certain time delay between the infrared light adjustment and the infrared camera image acquisition, the collected single vein image may have local vein features that are not obvious or lost

Moreover, due to the acquisition conditions and the different periods of the people being collected, the collected vein images will have certain differences, and the proportion of veins to the entire image is relatively small. Using a single vein image for vein feature extraction may cause feature instability or fewer features, thus affecting the accuracy of vein feature recognition

Images