PhD Computer Science Candidate Mr. Laurice Phillips on his thesis entitled “A Novel Approach to fingerprint Classification Using Regular Expression Machine Learning”.
The Department of Computing and Information Technology invites you to an oral examination presented by PhD Computer Science Candidate Mr. Laurice Phillips on his thesis entitled “A Novel Approach to fingerprint Classification Using Regular Expression Machine Learning”.
The presentation takes place on Monday 22nd January 2018, at 1:00 pm in the Department of Computing and Information Technology (DCIT), Conference Room.
Abstract: Biometric Security, Digital Image Processing and Machine Learning are hot topics in the field of computer science and information technology with a broad range of applications in industry, science and engineering. This research presents the development of a new approach to automated digital fingerprint classification. Firstly, a hybrid image enhancements strategy is employed to enhance the fingerprint image followed by the implementation of a novel technique for extracting the dominant singularity of the image. Finally, a new prediction model is trained to classify the digital fingerprint image using Regular Expression Machine system. This approach has never been attempted in the literature and this is supported by an international patent search report from the WIPO in Geneva Switzerland. Experimental results have shown an average accuracy of 93.78 % with some database subsets achieving as much as 97% accuracy.
