An Improved Algorithm And Study Of Fingerprint Identification System

With the extensive application of computer technology and the rapiddevelopment of information technology, computer technology for personalidentification has long-term development. Biometric technology has been selectedpriority action for identification tool, it is known that is closely integrated biometrictechnology through computer and optics, acoustics, biosensors and bio-statisticaltheory and other high-tech means, take advantage of the inherent physiologicalcharacteristics of the human body,(such as fingerprint, face like, iris, etc.) andbehavioral characteristics to identify personal identity, this paper introduces thefingerprint recognition, because in today’s society, fingerprint identification has itsown advantages and uniqueness of fingerprints, stability and anti-deceptive, making itone of today’s major identification techniques. However, it can be done not unilateral,it needs a lot of techniques including, for example, pattern recognition technology,digital image processing, artificial intelligence and other fields, the generation offingerprint recognition technology on our lives plays a big meaning, for example,banks systems, access control systems, the investigation of criminal cases, etc.,which require the application of fingerprint identification system to work is anintegral part of our lives, but it also gives us a lot to generate value and convenience.Fingerprint recognition technology including fingerprint preprocessing, featureextraction and fingerprint matching these three areas. Fingerprint imagepreprocessing, including fingerprint image normalization, image segmentation,calculated pattern, image enhancement, image binarization and thinning. The purposeof fingerprint image enhancement is to avoid generating pseudo-image featureinformation extracted minutiae ensure the accuracy and reliability. Because ofblockiness traditional Gabor filter based fingerprint image enhancement algorithm,this paper presents an improved fingerprint image enhancement algorithms. Theexperimental results show that the algorithm can be very good to remove the blockingeffect, making the process much smoother edge fingerprint images. Fingerprintcharacteristic extraction section, the fingerprint image due to various reasonsgenerates the false feature points, so we must remove the false feature points, thesubsequent step of our fingerprint matching has a critical influence. In the last article we introduced several algorithms related to fingerprint matching and proposes a newimproved algorithm to help us better optimize fingerprint system.