شبيه سازي يك سيستم تاييد امضاي برخط براي تشخيص اتوماتيك امضاي جعلي

چكيده انگليسي :

An Online Signature Verification System Simulated for Automatic Detection of Forgery Signature Mohsen Saeidi saeidim@ec iut ac ir Date of Submission April 29 2009 Department of Electrical and computer engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language FarsiSupervisors Rasoul Amirfattahi Fattahi@cc iut ac ir Mohammad Reza Ahmadzadeh Ahmadzadeh@cc iut ac irAbstract Biometrics is the utilization of biological characteristics face iris fingerprint or behavioral traits signature voice for identity verification of an individual Biometric authentication is gaining popularity as a more trustablealternative to password based security systems as it is relatively hard to be forgotten stolen or guessed Signature isa behavioral biometric it is not based on the physical properties such as fingerprint or face of the individual butbehavioral ones As such one s signature may change over time and it is not nearly as unique or difficult to forge asiris patterns or fingerprints however signature s widespread acceptance by the public make it more suitable forcertain lower security authentication needs Signature verification is split into two according to the available data inthe input Off line signature verification takes as input the image of a signature and is useful in automaticverification of signatures found on bank checks and documents On line signature verification uses signatures thatare captured by pressure sensitive tablets and could be used in real time applications like credit card transactions orresource accesses In this work we present a complete system for on line signature verification During registrationto system the user has to submit a number of reference signatures which are cross aligned to extract statisticsdescribing the variation in the user s signatures A test signature s authenticity is established by first aligning it witheach reference signature of the claimed user resulting in a number of dissimilarity scores distances to nearest farthest and template reference signatures In previous systems only one of these distances typically the distance tothe nearest reference signature or the distance to a template signature was chosen in an ad hoc manner to classifythe signature as genuine or forgery Here we propose a method to utilize all of these distances treating them asfeatures in a two class classification problem using standard pattern classification techniques The distances are firstnormalized resulting in a three dimensional space where genuine and forgery signature distributions are wellseparated We experimented with the Bayes classifier Support Vector Machines and a linear classifier used inconjunction with Principal Component Analysis to classify a given signature into one of the two classes forgery orgenuine In this research a technique to make a differ between genuine and forgery signature based on extremummatching for equalizing signal length is proposed In addition in three tank benchmark fault is detected by usingsupport vector machine also it is shown than support vector machin has better performance instead of fault in threetank benchmark detected and a comparison between this method and other method such as neural network radialbase function and back propagation been made In this research after doing a massive literature on support vectormachine optimization equation for classification we explain linear and nonlinear support vector machine with moredetail and with extreme simulation we investigate such parameter roll in fault detection and isolation Key words Support Vector Machine Dynamic Time Warping Extremum Matching Principle Component Analysis

سعيدي، محسن

شبيه سازي يك سيستم تاييد امضاي برخط براي تشخيص اتوماتيك امضاي جعلي

ماشين بردار پشتيبان , پيچش زماني پويا , تطابق نقاط اكسترم , آناليز اجزاي اصلي , PCA , كلوني مورچه ها