Fingerprint and finger vein image fusion method based on weighted fusion and layered serial structure

The invention provides a fingerprint and finger vein image fusion method based on weighted fusion and a layered serial structure, and mainly solves the problem that a single-mode system tends to be influenced by factors as noise and feature damage and a multi-mode fusion system is long in consumption time. Two designed fingerprint classifiers and four designed finger vein classifiers are trained, and identification rejection rates of the classifiers are obtained; the weighted a single classifier when the classifier is integrated in a fusion system is calculated according to the identification rejection rate; and to reduce operation time of an algorithm, a serial three-layer layered form is constructed for the six fused classifiers, samples cannot enter the second layer of classifiers unless passing the first layer of classifiers, and the samples cannot enter the third layer unless passing the second layer. According to the invention, fingerprint and finger vein images are identified via a plurality of feature