Fingerprint Recognition Technology


Fingerprint Recognition

FingerCode Fingerprint Recognition System

Neural Network Fingerprint Recognition

AOV Based Fingerprint Minutiae Matching System

Correlation Filters AFIS

Fingerprint Classification

Fingerprint Classification System

Singular Point Detection

Core Point Detection System

External resources

Advanced Source Code .Com

Neural Networks .It

Face Recognition .It

Iris Recognition .It

Image Compression .It

Speech Recognition .It

Speaker Recognition .It

Biometrics is method of recognizing a person based on a physiological or behavioral characteristic. Biometric system includes face, fingerprints matching, hand geometry, handwriting, iris, retinal, palm print, vein, and voice. This technology has become the foundation of highly secure identification and verification solutions. As the level of security gap and unfair transaction increases, then the need for secure identification and personal verification also increases. Biometric-based solutions provide confidential financial transactions and data privacy. The need for biometrics can be more in federal, state and local governments, military, and in commercial applications also. Network security, government ID proof, E-banking, and other money transactions, retail sales, and social services are already in benefit due to this technology. Biometric applications include network, domain access, single sign-on, logon, and data protection, access to resources, and transaction security. In biometric system iris and fingerprint technologies are widely used system as these two modalities are most reliable and possess uniqueness. Identification of fingerprint is most popular due to its unique characteristics which are used in application for identification of person. Fingerprint identification technology is a kind of important biological identity recognition technology; it is also the most mature branch of biological recognition technology development. Comparing with other biometric identification technology, fingerprint identification technology has high efficiency, low cost, convenient collection etc advantages. Now the development of the fingerprint recognition system becomes faster and faster, in order to meet the needs of the society, the development and application of embedded system also should follow closely the development direction of fingerprint identification technology to optimize the fingerprint identification algorithm, and develop with high recognition rate, fast processing speed, good scalability and low cost embedded platform, which has a broad market prospect and research value.

A fingerprint may be resolved into group of patterns bearing certain characteristics. A fingerprint consists of dark and light lines. The dark lines are the ridges while the lighter lines are referred as the valleys. The points where a ridge breaks into two are known as bifurcations and the points where a ridge discontinues are called the ridge ending. These points together are known as minutiae which distinctively and uniquely identify the fingerprint. Most automatic systems for fingerprint comparison are based on minutiae matching. The fingerprint verification schemes are either automated or in some cases two prints can be matched manually. Manual matching of prints is a tedious and time consuming activity. Human fingerprint examiners, in order to claim that two fingerprints are from the same finger evaluate several factors: global pattern configuration agreement, which means that two fingerprints must be of the same type, qualitative concordance, which requires that the corresponding minute details must be identical, quantitative factor, which specifies that at least certain number of corresponding minute details must be found, and corresponding minute details, which must be identically inter-related.

Fingerprint classification is the process of dividing a large amount of fingerprint database in which the input fingerprint is first determined and then a classification is carried out to observe the set of same class. A database usually contains a number of fingerprints with different number of individual features. The identification of input fingerprint within this database becomes an extremely long process. Therefore classification of fingerprint can help to increase the speed of identification. The input fingerprint is classified among the set of classes of fingerprint database. Thus each fingerprint is only need to match against the corresponding class contained in database. Fingerprint classification identifies the typical global representations of fingerprints. Global representations include locations of points (e.g., core and delta) in a fingerprint. A typical fingerprint classification is categories into the following six classes: whorl, right loop, left loop, arch, twin loop, and tented arch. It also contains one or more regions where the ridge lines consume different shapes (curvature or termination).

Fingerprint Recognition . It Luigi Rosa mobile +39 3207214179