Wavelet decomposition 4-feature parallel fusion by Quaternion Euclidean Product Distance matching score for palmprint verification

Parallel fusion is conventionally defined as a form of complex number for 2-feature. However, it is tough to represent 4-feature parallel fusion due to shortage of mathematical significance. To solve this problem, this paper proposes a reasonable interpretation by introducing quaternion. After parallel fusion whose features are extracted by wavelets decomposition coefficients from the same pixel corresponding to 4 separate sub-images, Quaternion Euclidean Product Distance (QEPD), a distance between modulus square of template quaternion and modulus of tester Quaternion Euclidean Product (QEP), as matching score, is performed. The scores discriminate between genuine and impostor of palmprint by threshold effectively. The approach saves the cost of computational time without training phase than existing algorithms and the experimental result shows a good recognition rate nearly achieves them but with a fast speed.