What can the structure of the palmprint tell us

Abstract

As this system is to be offered as an online palmistry service, the identification process of the palmprint has to be fast and accurate. Therefore, the purpose of this project is to review on the accuracy of the existing palmprint identification system, developed by previous students using pairwise angle difference and subset divisions, and improve the time complexity. These were done by reducing any redundancy through optimization of code and implementation of Subset Reordering of Combine Histogram before performing Discrete Fourier Transform (DFT) on the subsets, as well as demonstrating the feasibility of the proposed Integrated Discrete Fourier Transform – Combine Histogram Algorithm (IDFT- CH) to enhance performance of the system.

The results gathered from the simulations of the optimized code improved the overall system’s time complexity by 40% and the Subset Reordering implementation was found to be able to further reduced the time complexity of the proposed IDFT-CH system by 40% as compared to the 20% reduction without the Subset Reordering implementation.

Kewords: Palmprint Identification, Pairwise Angle Difference, Subset Divisions, Subset Reordering, Discrete Fourier Transform, Combine Histogram