CMPE 571
BİIMETRIC IDENTIFICATION AND VERIFICATION SYSTEMS
HOMEWORK -1
In the attached files you will find 30 fingerprint image pairs (30 images for training and 30 images for testing, for each finger there is one train image and one test image. For example, f001_01.png is the train image, and s001_01.png is the testing image). For the given images design a biometric authentication system that will use LBP (Local Binary Pattern) as the feature extractor. You may use one of the classifiers (kNN, NN, SVM, Naive Bayes, Decision tree, etc).
Some notes about the homework:
• Please briefly describe your methodoly
• Please calculate the performance of the system
• Please write your comments clearly about the design and performance
• BONUS: You may use SURF as a feature extractor
• BONUS: You may try different classfiers