International Journal of Emerging Technology in Computer Science and Electronics ( IJETCSE)
Pattern classification can be characterized as "The demonstration of taking in crude information and making a move in light of the gathering of the method". Mostly the pattern classification techniques are utilized as a part of security applications, for example, biometrics based individual recognition and interruption recognition in PC networks and spam filtering, goal to segregate between the honest to goodness and a malicious pattern class. Biometric tasks are different from classical pattern recognition tasks Biometric spoof attack using fake fingerprints. Since the pattern classification techniques do not take into account the adversarial nature of recognition, here exhibit performance degradation when used in adversarial setting, namely under attacks.