Hybrid model for biometrics based Human identification and recognition System using hand dorsum geometry And finger knuckle print

Abstract

The adoption of Biometrics based Automatic Human Recognition newlineand Identification System is gaining importance and more focus Reliability in newlinethe human recognition is the key to the security Many physiological newlinecharacteristics of human biometrics are typically time invariant easy to newlineacquire and unique for every individual Biometric features such as face iris newlinefingerprint hand geometry palmprint signature etc have been suggested for newlinethe security in access control Most of the previous researches in biometrics newlinehave been focused on fingerprint and face The reliability of personal newlineidentification using face is low due to the problems of pose lighting newlineorientation and gesture Fingerprint identification is widely used in personal newlineidentification as it has worked well in most cases in the past However it is newlinedifficult to acquire fingerprint features minutiae for certain class of newlinepersons such as manual labourers elderly people etc As a result other newlinebiometric characteristics are receiving increasing attention newlineBiometric systems that operate using any single biometric newlinecharacteristic have some limitations which may lead to poor identification newlineresults due to many reasons such as noise in sensed data Intra class newlinevariations distinctiveness non universality spoof attacks etc Some of the newlinelimitations imposed by unimodel biometric systems can be overcome by using newlinemultiple biometric modalities newline newline