Comparison of Classification methods for perspiration-based liveness algorithm

دانشگاه آزاد اسلامی اصفهان(خوراسگان)

منو

رکورد قبلی

رکورد بعدی

" Comparison of Classification methods for perspiration-based liveness algorithm "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	151963
Doc. No	:	ET23755
Main Entry	:	Sujan T.V. Parthasaradhi
Title Proper	:	Comparison of Classification methods for perspiration-based liveness algorithm
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	In the modern world there is a need for security. Biometric technologies provide ameans for providing this security. Of the many different available biometric technologies,fingerprint recognition is the most popular. As with all security measures, biometricdevices may be subject to attacks on the system. Fingerprint scanners may be susceptibleto spoofing using artificial materials, or in the worst case, dismembered fingers. Liveness,i.e. to determine whether the introduced biometric is coming from a live source, has beensuggested as a means to circumvent attacks that use spoof fingers. It has been shown thatwater based casting materials and cadaver fingers were able to be scanned and verifiedfor most fingerprint scanner technologies. In our laboratory an anti-spoofing methodbased on liveness detection has been developed for use in fingerprint scanners. Thismethod quantifies a specific temporal perspiration pattern present in fingerprints acquiredfrom live claimants. For this thesis, perspiration detection algorithm is optimized fordifferent fingerprint scanner technologies, using a larger, more diverse data set, and ashorter time window. Several classification methods are tested in order to separate liveand spoof fingerprint images. Each method had a different performance with respect toeach scanner and time window. All the classifiers achieved approximately 90 classification rate for all scanners, using the reduced time window and the morecomprehensive training and test sets. Based on the classification results, it is believed thatthis perspiration-based method has a potential to reduce the susceptibility of thefingerprint scanners to spoof attacks..............-....,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL47013.pdf
Title and statement of responsibility and	:	Comparison of Classification methods for perspiration-based liveness algorithm [Thesis]