Application of Polar Harmonic Transforms in Thumb Impression Recognition
Prakash Choudhary1, Neha Mahala2, Khusboo Uprety3, P K Bhagat4
1Prakash Choudhary, National Institute of Technology Manipur, India.
2Neha Mahala, ISM Dhanbad, India. Khusboo Uprety, National Institute of Technology Manipur, India.
3P K Bhagat, National Institute of Technology Manipur, India.
Manuscript received on 10 October 2017 | Revised Manuscript received on 18 October 2017 | Manuscript Published on 30 October 2017 | PP: 137-141 | Volume-7 Issue-1, October 2017 | Retrieval Number: A5217107117/17©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Fingerprint recognition refers to the methods of matching or verifying a known and questioned fingerprint against another fingerprint to ascertain if the impressions are the same. Fingerprints are the most popular biometrics to authenticate a person as it is unique and permanent throughout a person’s life. Polar Harmonic Transforms (PHTs) are orthogonal rotation invariant 2D transforms that provide various numerically stable features for fingerprint recognition. The kernel functions of PHTs are basic waves and harmonic in nature that consists sinusoidal functions that are inherently computation intensive that can be used to generate rotation invariant features. PHTs are characterized by low time complexity and numerical stability. In this paper, Polar Harmonic Transforms (PHTs) are introduced for rotation invariance in thumb impression recognition, namely, Polar Complex Exponential Transform (PCET), Polar Cosine Transform (PCT), and Polar Sine Transform (PST). Orthogonal kernels of PHTs are more effective in terms of information compactness and minimal information redundancy. A fast approach of computation of Polar Harmonic Transform for thumb impression recognition with low values of FAR and FRR have been implemented. The accuracy obtained is above 80 percent.
Keywords: Fingerprint, PCET, PCT, PHT, PST, Rotation Invariant PHT,
Scope of the Article: Pattern Recognition