Parallel algorithms for iris biometrics

Abstract

Iris biometrics involves preprocessing, feature extraction and identification phase. In this thesis,an effort has been made to introduce parallelism in feature extraction and identification phases. Local features invariant to scale, rotation, illumination are extracted using Scale Invariant Feature Transform (SIFT). In order to achieve speedup during feature extraction, parallelism has been introduced during scale space construction using SIMD hypercube. The parallel time complexity is O(N2) whereas sequential algorithm performs with complexity of O(lsN2, where l is the number of octaves, s is the number of Gaussian scale levels within an octave and N × N is the
size of iris image.