Signature Recognition using 2D Discrete Wavelet Transforms
Meenu Kumari1, Anil Kumar2, Manish Saxena3
1Meenu Kumari*, Department of Physics, IFTM University, Moradabad, India.
2Anil Kumar, Department of Physics, Hindu College, Moradabad, India.
3Manish Saxena, Department of Physics, Moradabad Institute of Technology, Moradabad, India.
Manuscript received on April 20, 2020. | Revised Manuscript received on May 01, 2020. | Manuscript published on May 10, 2020. | PP: 528-532 | Volume-9 Issue-7, May 2020. | Retrieval Number: F4399049620/2020©BEIESP | DOI: 10.35940/ijitee.F4399.059720
Open Access | Ethics and Policies | Cite | Mendeley
© 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: Signature recognition is one of the most secure techniques used for person identification. Wavelet transforms possess an extra time localization property over Fourier transforms, despite both are frequency localized. That is why; it is more useful to analyze the one dimensional and two dimensional signals both. First of all the features of a signature are extracted and then matching is performed. We have proposed feature extraction technique using discrete wavelet transforms level-1 and matching of signatures are performed using some statistical parameters of discrete wavelet coefficients using matching percentage, sum of absolute difference, mean square error and city block distance.
Keywords: Signature recognition, discrete wavelet transform, matching percentage, sum of absolute difference and mean square error, city block distance.
Scope of the Article: Discrete Optimization