Finger vein recognition method and system based on multi-source domain migration

Abstract

The invention provides a finger vein recognition method based on multi-source domain migration, and the method comprises the following steps: firstly, obtaining an ROI corresponding to a finger vein image of a target domain; inputting the ROI into a basic feature extraction network to extract basic features; respectively inputting the basic features into the CFTN and the DFTN to obtain general features and domain specified features; splicing the general feature and the domain specified feature to obtain a final aggregation feature; searching the obtained aggregated features in an existing finger vein feature database to obtain an input finger vein picture and a matching score of the existing finger veins in the database; and finally, outputting a matching result of the input finger vein according to the matching score. By migrating a plurality of source domains to a target domain, the sample demand of the target domain is reduced, so that the acquisition and labeling cost is reduced; and universal knowledge of multiple source domain data sets is migrated to a target domain, and domain specified features are reserved, so that the performance of migration learning is improved to the greatest extent.

Description

technical field [0001] The invention belongs to the field of biological feature recognition, in particular to a finger vein recognition method and system based on multi-source domain migration. Background technique [0002] Finger vein recognition is a newly proposed modality for biometric recognition that is user-friendly, highly secure, and possesses natural anti-counterfeiting properties. More and more researchers and engineers have begun to devote themselves to research in the field of finger vein recognition. However, in the practice of finger vein engineering, finger vein recognition is still affected by the following two problems: [0003] Image difference: Due to the difference between the optical imaging sensor and near-infrared light, the finger vein images collected by different devices in different scenarios are different, such as figure 1 shown. In addition, their grayscale histograms and local binary coded images are also listed below to illustrate their dif...

AI Technical Summary

Problems solved by technology

c. Generator gradient disappearance problem: When the discriminator is very accurate, the loss of the discriminator quickly converges to 0, so that it cannot provide a reliable path to continue to update the gradient of the generator, causing the gradient of the generator to disappear

The training of GAN is due to the random noise distribution at the beginning, which is too far away from the real data distribution, and there is almost no overlap between the two distributions. At this time, the discriminator can quickly learn to distinguish the real data from the generated fake data. To achieve the optimum of the discriminator, the gradient of the generator cannot continue to be updated or even the gradient disappears

That is, the domain adaptive method and data amplification method based on GAN are relatively unstable

Images