dcit_seminar_flyer_header_title_wbackground.jpg

 

Presents: Mr. Laurice Phillips

1st Ph.D. Research Seminar

  DCIT_Seminar_PhD_Laurice_Phillips_1_1_image.jpg  

 

Title: Machine Learning Algorithm For Automatic Fingerprint Image Classification And Recognition

DATE: Thursday 22nd November 2012

TIME:2:00 p.m.

VENUE: Department's Seminar room, Dept. of Computing and Information Technology, 2nd floor, Natural Sciences Building

 

ABSTRACT: In today’s society many biometric systems use automatic fingerprint classification and recognition. Many of the systems proposed used techniques such as fuzzy logic, neural network, supervised, unsupervised learning algorithms or a combination of techniques. Nevertheless, the problem of automatic fingerprint identification still remains open. Several complex structures, algorithms and techniques have been developed to provide solutions that have been successfully implemented in fingerprint systems. Generally all systems employ some measure of image enhancement as one of its crucial steps. 

In recent times Support Vector Machine (SVM) learning algorithms have gained popularity as a useful methodology. In fact, SVM theory has been already implemented successfully in image processing related systems. This research seeks to apply SVM theory to the problem of automatic fingerprint identification.  The research, in part, demonstrates an application written in the Java platform that is used as a basis for the design, implementation and experimentation of the SVM algorithms and methods used in our proposed fingerprint system. There are many methods used in the development of fingerprint image enhancement. This research seeks to uncover a methodology that delivers significant improvement over the existing approaches used in addressing the problem of automatic fingerprint identification systems.

Keywords: Image Enhancement, Classification, Support Vector Machine, Supervised Learning, Automatic Fingerprint identification.

 

For more information Click here

 

Top of Page