face recognition

Face Recognition Technology for User Authentication and Proactive Surveillance:

Face Recognition Technology for User Authentication and Proactive Surveillance BY:- INDRESH CHATURVEDI Mca ii year I.T.S Mohan Nagar GZB

Biometrics : Face recognition:

Biometrics : Face recognition Biometrics refers broad range of technologies based on human characteristics. Physiological : Face, fingerprint, Iris, DNA. Behavioral : Hand-written signature, voice. Characteristics 011001010010101… 011010100100110… 001100010010010... Templates

Classification of biometric traits :

Classification of biometric traits BIOMETRICS PHYSIOLOGICAL BEHAVIORAL IRIS FINGER PRINT HAND FACE DNA VOICE SIGNATURE KEY STROKE

Three Basic Identification Methods:

Three Basic Identification Methods Password PIN Keys Passport Smart Card Face Fingerprint Iris Universal Unique Permanent Collectable Acceptance Universal Unique Permanent Collectable Acceptance Universal Unique Permanent Collectable Acceptance Possession (“something I have”) Biometrics (“something I am”) Knowledge (“something I know”) “sanjay” “750426” ? ü ü ü ü ü ü ü ü

Face Recognition : Procedure :

Face Recognition : Procedure Enrollment Test Verification SENSOR Pre Processing Feature Extractor Template Generator Matcher Stored Templates Application device

Identification vs. Verification:

g Identification (1:N) Biometric reader Biometric Matcher Identification vs. Verification Image Database Verification (1:1) Biometric reader Biometric Matcher ID Image Database This person is xyz Match I am xyz Enrollment subsystem Authentication subsystem

TECHNOLOGY TREND:

TECHNOLOGY TREND Three matching methods: Feature-based (structural) matching : find the location of eyes, nose & mouth ,extract the feature point . And also use distance between eyes corner & angle between eyes corner . Person image pointed image

Holistic matching : eigenface:

Holistic matching : eigenface Decompose face images into a small set of characteristic feature images. A new face is compared to these stored images. A match is found if the new faces is close to one of these images. Training set eigenfaces

Neural Networks & TS-SOM:

Neural Networks & TS-SOM Individual units to simulate Neurons Parallel Processing Many inputs and single output

TS-SOM :

TS-SOM Tree structure self-organizing maps Each unit of map receives identical inputs Units complete for selection

NEURAL NETWORK PROCESS ;:

NEURAL NETWORK PROCESS ;

Known limitations ::

Known limitations : Lighting and angle can affect performance Range can affect the performance Biometric solutions are close to 100%, but not 100%, there could still be false acceptance and false rejection .

FUTURE DEVELOPEMENT:

FUTURE DEVELOPEMENT Mobile authentication ( Application in mobile phone) IR-based technology ( To achieve excellent accuracy) 3D face recognition ( Under research)

APPLICATIONS OF FACE RECOGNITION :

APPLICATIONS OF FACE RECOGNITION Verification of credit card, personal ID, passport Access control system Human-computer-interaction verifications for criminals persons

Questions and comments:

Questions and comments Thank you for your Attention!