Recognition Technology Research Based On 3D Fingerprint

In recent years, it is increasing emphasis on the safety of identification. Biometric features recognition has many advantages, such as security, confidentiality, and convenience. Therefore biometric technology has been widely applied to various areas in life. Fingerprint, as a unique, changeless and easily acquired biometric, has been widely applied to border control, access control, criminal investigation, and so on. Fingerprint has been the most important biometrics. However, current used fingerprint identification technology is essentially based on contact collected fingerprint image. Due to the difference of the intensity and perspectives, this push-acquisition mode, deforms the obtained fingerprints. Therefore, identification accuracy will be greatly affected.This paper presents a method based on the three-dimensional(3D) fingerprint recognition technology. Due to the non-contact acquisition mode, 3D fringe projection measurement technology has no distortion problem and the characteristics of the obtained fingerprint information are greater than two-dimensional(2D) fingerprint. Because the current 2D fingerprint database is popular and irreplaceable, the collected 3D fingerprints are unrolled to a 2D equivalent fingerprints in order to make a 3D and 2D fingerprint interoperable. Calculating the fingerprint orientation field obtains the essential texture of fingerprint and then binarizes them. The fingerprint image is converted to black and white images. Thinning is a procedure to make the ridge line have one pixel width in the binary image, which makes it easier extract the fingerprint minutiae points. The extracted minutiae generate a database. If the matching minutiae exceed the setting threshold value, two fingerprints are matched. Experimental results on measuring and recognizing some 3D fingerprints show the accuracy and availability of the developed 3D fingerprint system.