False fingerprint detection method based on finger wave conversion and SVM

Abstract

The invention discloses a false fingerprint detection method based on finger wave conversion and a support vector machine (SVM). The method includes the first step of collecting a true fingerprint image and a false fingerprint image, the second step of carrying out finger wave conversion on the collected true fingerprint image and the collected false fingerprint image, the third step of extracting finger wave conversion parameters of the collected fingerprint images as feature vectors, the fourth step of carrying out normalization processing on the extracted feature vectors, the fifth step of carrying out SVM training on the normalized feature vectors to obtain a classifier based on the SVM, and the sixth step of classifying the normalized feature vectors of fingerprints to be detected through an SVM classification model obtained in the fifth step so as to obtain an SVM classification result which shows whether fingerprints are true or false. The feature vectors are extracted after finger wave conversion is carried out on the fingerprint images, and input into the SVM for classification, and true fingerprints and false fingerprints can be well distinguished on the condition of not changing hardware of a fingerprint collection instrument. Experimental results show that the false fingerprint detection method is high in accuracy, low in false rejection rate and false accept rate, low in cost and good in practicability.

Description

technical field [0001] The design of the invention belongs to the field of image processing and pattern recognition, relates to a false fingerprint detection method, in particular to a false fingerprint detection method based on finger wave transformation and SVM. Background technique [0002] At present, fingerprint identification technology is widely used in various identification systems, and it has good advantages in terms of identification accuracy and operation convenience. [0003] However, many criminals use various materials to forge fingerprints to pass through the fingerprint identification system. True fingerprints are also called living fingerprints, which refer to the fingerprints of fingers with human biological characteristics, that is, the fingerprints of living human bodies. The corresponding fake fingerprints are also called dead body fingerprints, including fingerprints made of various materials, such as silica gel, clay, paper with printed fingerprint i...

AI Technical Summary

Problems solved by technology

However, in the current method of distinguishing true and false fingerprints, only some simple statistical characteristics and frequency domain characteristics are extracted. The method is too simple and cannot effectively distinguish true and false fingerprints.

Images