Fingerprint local invariant feature extraction on GPU with CUDA

Driven from its uniqueness, immutability, acceptability, and low cost, fingerprint is in a forefront between biometric traits. Recently, the GPU has been considered as a promising parallel processing technology due to its high performance computing, commodity, and availability. Fingerprint authentication is keep growing, and includes the deployment of many image processing and computer vision algorithms. This paper introduces the fingerprint local invariant feature extraction using two dominant detectors, namely SIFT and SURF, which are running on the CPU and the GPU. The paper focuses on the consumed time as an important factor for fingerprint identification. The experimental results show that the GPU implementations produce promising behaviors for both SIFT and SURF compared to the CPU one. Moreover, the SURF feature detector provides shorter processing time compared to the SIFT CPU and GPU implementations.