3617

3417

كارشناسي ارشد

الكترونيك

اصفهان :دانشگاه صنعتي اصفهان، دانشكده برق وكامپيوتر

1386

ده ، 144، [II]ص.: مصور، جدول ،نمودار

ص. ع. به فارسي وانگليسي

مهندسي برق و كامپيوتر

ID3417

به فارسي وانگليسي : قابل رويت در نسخه ديجيتال

Abstract Nowadays security is one of the most important challenges in the life of any modernhuman being One of the means of providing security is through identification of people To provide identification science of biometrics has been employed Biometrics is a sciencethat uses the behavioral and biological characteristics of a person to identify him Fingerprint verification is an important biometric technique for personal identification Most of the automatic verification systems are based on matching of fingerprint minutiae Extraction of minutiae is an essential process in design of fingerprint verification systems Estimation of orientation of the lines in an image is one of the complicated processes forenhancement of fingerprint and minutiae extraction In this thesis with improvement andimplementation of existing algorithms a hardware scheme is presented which is based on apipelined architecture in order to perform real time orientation estimation Most of the minutiae extraction methods use binarization to convert the grayscaleimage into a binary image Recently a number of methods have been devised to extract theminutiae directly from the grayscale image These routines require complex and timeconsuming computations and hence are performed through software This complexity hashindered their hardware implementation In this thesis a new approach is offered whichsignificantly reduces the required computations and extracts the minutiae from the imagewith relatively high speed Also offered is a new architecture which implements thesuggested method

شاهرخ سماوي

سعيد صدري

مسعود سيدي، رسول دهقاني