Signature Verification
Using CNN
Submitted By:
Rachana Kulkarni (3747)
Sayali Kawale(3753)
Rifa Shazmeen(3755)
Introduction
 Computers are largely used in almost each and every field including financial
transactions, thus providing additional security measures for computer system is
necessary.
 The security measures to be used must be cheap, reliable and un-intrusive to the
authorized person. The technique which meets these requirements is handwritten
signature verification.
Why Handwritten Signature only ?
 Everyday transaction
 Individuals less likely to object
 Individual’s signature is unique
 Less controversial
Importance of Signature
 Transaction
 Individuals less likely to object
 Biometric
Types of Signature Identification
 Offline Signature Verification
 Deals with shape only
 Online Signature Verification
 Deals with dynamic features like speed, pen pressure, directions, stroke length, and
when the pen is lifted from the paper
Challenging tasks
 The main challenge with online Signature Identification technology is
differentiating between the consistent parts of the signature and the behavioural
parts of the signature that vary with each signing.
 A person’s signature is never entirely the same. Allowing for these variations while
providing protection against forgers is a challenging task.
 Differentiating between the parts of the signature that vary with each signing.
 The signature can vary substantially over an individual’s lifetime.
Proposed Work
 Signature Acquisition
 Signature Pre-processing
 Feature Extraction
 Signature Verification
Signature Acquisition
 Signature is acquired from the user
Signature Pre-processing
 Function of the pre-processor
• Image resizing
• Image binarizing
• Image thinning
• Image normalizing
Feature Extraction
Signature Identification
Applications
 Banking
 Passport office
 And any other places which require identification !
Advantages
 Low error rate
 Forgery is detected even when the forger has managed to get a copy of the
authentic signature
 Fast and simple training
 Cheap hardware
 Little storage requirements
Activity Diagram
Requirements & Technologies
 The hardware component we are using here is a scanning device(WEBCAM), high
RAM for better results and good processor.
 Operating System : Windows XP or Higher
 NetBeans IDE 7.0
 SDK – J2SE
 Intel core 2 duo processor
 2.1 GHZ, 1 GB RAM
Signature Verification.pptx

Signature Verification.pptx

  • 1.
    Signature Verification Using CNN SubmittedBy: Rachana Kulkarni (3747) Sayali Kawale(3753) Rifa Shazmeen(3755)
  • 2.
    Introduction  Computers arelargely used in almost each and every field including financial transactions, thus providing additional security measures for computer system is necessary.  The security measures to be used must be cheap, reliable and un-intrusive to the authorized person. The technique which meets these requirements is handwritten signature verification.
  • 3.
    Why Handwritten Signatureonly ?  Everyday transaction  Individuals less likely to object  Individual’s signature is unique  Less controversial
  • 4.
    Importance of Signature Transaction  Individuals less likely to object  Biometric
  • 5.
    Types of SignatureIdentification  Offline Signature Verification  Deals with shape only  Online Signature Verification  Deals with dynamic features like speed, pen pressure, directions, stroke length, and when the pen is lifted from the paper
  • 6.
    Challenging tasks  Themain challenge with online Signature Identification technology is differentiating between the consistent parts of the signature and the behavioural parts of the signature that vary with each signing.  A person’s signature is never entirely the same. Allowing for these variations while providing protection against forgers is a challenging task.  Differentiating between the parts of the signature that vary with each signing.  The signature can vary substantially over an individual’s lifetime.
  • 7.
    Proposed Work  SignatureAcquisition  Signature Pre-processing  Feature Extraction  Signature Verification
  • 8.
    Signature Acquisition  Signatureis acquired from the user
  • 9.
    Signature Pre-processing  Functionof the pre-processor • Image resizing • Image binarizing • Image thinning • Image normalizing
  • 10.
  • 11.
  • 12.
    Applications  Banking  Passportoffice  And any other places which require identification !
  • 13.
    Advantages  Low errorrate  Forgery is detected even when the forger has managed to get a copy of the authentic signature  Fast and simple training  Cheap hardware  Little storage requirements
  • 14.
  • 15.
    Requirements & Technologies The hardware component we are using here is a scanning device(WEBCAM), high RAM for better results and good processor.  Operating System : Windows XP or Higher  NetBeans IDE 7.0  SDK – J2SE  Intel core 2 duo processor  2.1 GHZ, 1 GB RAM