Evaluating Gabor preprocessing for SIFT-based ocular recognition
P. Campos, Micheal Forkin, Xiao Wen Xu · 2011
It has been shown that SIFT is an effective recognition algorithm when applied to the ocular region[2, 4]. In addition, Gabor filters have been shown to be very useful as a preprocessing tool for machine learning driven image classification[1]. We evaluate the effect of this preprocessing approach on the performance of the SIFT algorithm when applied to ocular recognition. Our results show that the Gabor filter preprocessing does not improve the performance of the SIFT algorithm in the realm of ocular recognition.