A multimodal biometric detection system via rotated histograms using hough lines

Fatma Susilawati, Mohamad and Zahraddeen, Sufyanu (2015) A multimodal biometric detection system via rotated histograms using hough lines. ARPN Journal of Engineering and Applied Sciences, 10 (3). pp. 1479-1485. ISSN 18196608

[img] Image
Restricted to Registered users only

Download (188kB) | Request a copy


Several systems require full identification of a user, as any misclassification may deteriorate the performance of the entire system. Such systems must grant access only to the genuine user. For this reason, single biometrics becomes insufficient for authentication and identification. Consequently, the need for implementing highly integrated systems is necessary to promote security of such systems. At the same time, multi-biometric attracts much attention. The current study put forward a pioneering multimodal biometric detection approach using the principle of detecting lines through Hough Transform (HT). The images were converted in to histograms using histogram plot function. However, these histograms images were rotated by 30 degrees and HT functions were applied on the rotated histograms to detect the query biometric features. The new technique was tested on face, iris, palm and fingerprint. The final plot accomplished detection of whole biometric features with an average detection time of 4.506 seconds per individual. The new technique can be used to detect the aforementioned biometric traits using the same feature extraction algorithm at limited time, since each biometric trait’s dimensions was drastically reduced. The new system outperformed many methods in the literature reported using conventional detection methods. Hence, the modified algorithm is applicable in multi-biometrics detection prior to recognition especially where little computation and fast performance is highly demanded.

Item Type: Article
Uncontrolled Keywords: histogram, multi-biometrics, HOUGH transform (HT)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Informatics & Computing
Depositing User: Syahmi Manaf
Date Deposited: 13 Sep 2022 05:50
Last Modified: 13 Sep 2022 05:50
URI: http://eprints.unisza.edu.my/id/eprint/5864

Actions (login required)

View Item View Item