Page 64 - Forensic News Journal Jan Feb 2018

Forensic News Journal
Finger Print Recognition Process and Parameters, pg 64-67
©2014-2018 SYT Global, Inc.

Finger Print Recognition

Process and Parameters

By Brindha Ramamoorthy

Introduction Automatic verification colored lines on the walls
methods based on physical of the caves at Lascaux,

In many business transac- biometric characteristics France, painted in prehis-
tions and access to privi- such as fingerprint or iris toric times, as mammals

leged information, reliable can provide positive veri- of prey. The patterns in
and accurate verification fication with a very high biological data contain
of people is extremely im- accuracy. However, the knowledge, if only we can

portant. Forensic science biometrics-based methods discover it. Discrimination
labs and identification assume that the physical of signal patterns allows

units for criminal investi- characteristics of an indi- personal identification by
gations routinely use fin- vidual used for verifica- voice, handwriting, fin-
gerprints. More recently tion are sufficiently unique gerprints, facial images,

an increasing number of to distinguish one person and so on, as well as the
civilian and commercial from another. Identical recognition of speech,

applications like welfare twins have the closest ge- written characters, and
disbursement, cellular, netics-based relationship scenes in images. It also
phone access, laptop com- and therefore the maxi- includes the identification

puter login are either us- mum similarities between of military targets based
ing or actively considering fingerprints are expected on radar, infrared, and/

to use fingerprint based to be found among identi- or video images. Patterns
verification because of the cal twins. exist in high-frequency
availability of inexpen- electromagnetic scans of

sive and compact solid Pattern Recognition body chemicals and other
state scanners as well as organic chemicals, includ-

its superior and proven The concept of pattern is ing DNA. The concept of
matching performance universal in intelligence classification involves the
over other biometric tech- and discovery. For ex- learning of likeness and

nologies. ample, we perceive the differences of patterns that

64