Fingerprint recognition is the process of the verification of a person’s identity by comparing their fingerprints with previously recorded samples. Fingerprints are impressions of human fingers, consisting of a series of ridges and grooves. Fingerprints captured in the system are located by minutiae points – points at which scars begin or terminate. These are further mapped with lines between them, creating a minutiae template.
Innovatrics fingerprint recognition algorithm consists of two main parts: feature extractor and fingerprint matcher.
Feature extractor takes the fingerprint image as a raw image and encodes it in a specific fingerprint template. Innovatrics’ extraction algorithm supports different image formats from various types of sensors. It is also optimized for inked and rolled images (commonly used in criminal applications).
The feature extractor was designed to work well with low quality and partial fingerprint images. Fingerprint image quality can be degraded due to less than optimal conditions during acquisition. The algorithm is able to considerably enhance the overall image quality and to fix possible defects in a way that these will not alter the recognition process. These advanced image enhancement techniques have a significant impact on the overall accuracy of the system.
The purpose of the matcher is to compare two fingerprint templates. The matcher doesn’t use the original fingerprint image but rather the resulting templates. It then produces a similarity score which indicates whether two fingerprint templates originate from the same finger or not.
Fingerprint verification is an identity authentication process used to confirm a claimed identity by comparing a captured template against a previously enrolled 1:1. When the templates match, verification is successful.
Innovatrics’ matching algorithm can also perform a high-speed identification search. Identification can be regarded as a generalization of verification; the goal of the identification process is to find a person in the database containing multiple identities (1:N search). The database size can be variable – from a few hundreds to tens of millions of templates can be stored in the database depending on the application.
Innovatrics’ fingerprint algorithms are among the best both in verification and identification according to independent NIST benchmark
The National Institute of Standards and Technology (NIST) is the leading international biometric testing organization and industry-standard developer. NIST is actively involved in developing architectures and conformance test tools to support users that need to comply with selected biometric standards and support product developers and testing laboratories interested in conforming to biometric standards by using the same testing tools available. NIST conducts technology evaluations of fingerprint matching, identification, and verification systems such as:
Biometric systems are used in a wide range of applications, where accurate identification or verification is needed. Modern use cases of fingerprint technology are: