특징점의 융선 형태정보를 이용한 지문인식
Fingerprint reconition using ridge shape of minuriae
융선형태정보 지문인식 제어계측공학 전산공학;
- 원문 URL
Recently, the social requirement of personal identification techniques has been expanding in new application areas. Fingerprint recognition is one of the biometrics methods available that has been widely used for that requirement. This paper proposes the fingerprint matching algorithm using the data of the ridge shapes directly connected with minutia. The data are expressed in one-dimensional discrete-time signals. Extracted minutia, which are classified into ridge bifurcations and ridge endings, are stored with their positions, types and ridge directions and then used for alignment and matching. The proposed algorithm is consist of 1)obtaining one-dimensional discrete-time signals at all minutiae from input fingerprint and register one, 2)finding pairs of minutia which have the similar ridge shape by comparing input fingerprint with register one, 3)finding rotation angle and moving displacement candidates used for the alignment from the pairs of similar minutia, 4)obtaining the final rotation angle and moving displacement value within those candidates set for the real alignment by using the Clustering method, 5)aligning an input fingerprint with a register one, 6)calculating the matching rate by counting the number of corresponded pairs of minutia within the overlapped area of an input and a registered fingerprint. As a result of experiment, false rejection rate(FRR) of 0.683 at false acceptance rate(FAR) of 0 is achieved.