An efficient fingerprint identification algorithm based on minutiae and invariant moment.

While we are experiencing many advantages of digital technologies and products, the security issues are also attracting increasing concerns. The secure fingerprint identification has become one of the most important research topics because of these increasing concerns. It has promoted us to propose an efficient fingerprint identification algorithm based on minutiae and invariant moment in this paper. In the proposed algorithm, the raw fingerprint image is first enhanced by the short-time Fourier transform (STFT). After that, the fingerprint minutiae can be extracted, which thereafter are selected as the center of region of interest (ROI), according to morphological transformation. Finally, a metric called cosine similarity among invariant moments is utilized to judge the similarities between fingerprint objects in the procedure of identification. The proposed algorithm is not limited to the fingerprint image with the fingerprint core. Experimental results show that the proposed algorithm provides a better performance in terms of matching accuracy when compared with the related works. The average accuracy is up to 96.67%. Moreover, the use of invariant moment of the ROI can avoid the leakage of fingerprint information and improve the security level of fingerprint recognition. Therefore, the proposed scheme is potentially used in many applications, such as smart fingerprint lock, intelligent community management information system, and automation control of home appliances. [ABSTRACT FROM AUTHOR]