Parallel scale space construction using SIMD hypercube

Author

Panda, A.C. ; Mehrotra, Harshit ; Majhi, Banshidhar

Author_Institution

Dept. of Comput. Sci. & Eng., Nat. Inst. of Technol. Rourkela, Rourkela, India

fYear

2012

fDate

Oct. 30 2012-Nov. 2 2012

Firstpage

625

Lastpage

628

Abstract

This paper proposes parallel scale space construction of Scale Invariant Feature Transform (SIFT) using SIMD hypercube. The parallel SIFT approach is used for iris feature extraction. The input iris images and Gaussian filters are mapped to each processor in the hypercube and convolution takes place in each processor concurrently. The time complexity of parallel algorithm 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 for N² sized iris image.

Keywords

Gaussian processes; computational complexity; feature extraction; iris recognition; parallel algorithms; transforms; Gaussian filters; Gaussian scale levels; N² sized iris image; SIMD hypercube; iris feature extraction; iris images; parallel SIFT approach; parallel algorithm; parallel scale space construction; scale invariant feature transform; sequential algorithm; time complexity; Convolution; Feature extraction; Hypercubes; Iris; Iris recognition; Kernel; Time complexity;

fLanguage

English

Publisher

ieee

Conference_Titel

Information and Communication Technologies (WICT), 2012 World Congress on

Conference_Location

Trivandrum

Print_ISBN

978-1-4673-4806-5

Type

conf

DOI

10.1109/WICT.2012.6409151

Filename

6409151