Reference point detection for camera-based fingerprint image based on wavelet transformation

• Published in: Biomedical engineering online Vol. 14; no. 1; p. 40
• Main Author: Khalil, Mohammed S
• Format: Journal Article
• Language: English
• Published: London BioMed Central 30.04.2015; BioMed Central Ltd
• Subjects: Analysis; Anatomic Landmarks - anatomy & histology; Anthropometry - instrumentation; Anthropometry - methods; Biomaterials; Biomedical Engineering and Bioengineering; Biomedical Engineering/Biotechnology; Biotechnology; Dermatoglyphics; Engineering; Fourier Analysis; Humans; Image Processing, Computer-Assisted - methods; Pattern Recognition, Automated - methods; Photography - instrumentation; Scanning devices; Signal-To-Noise Ratio; Smartphone; Wavelet Analysis; Wavelet; Mobile phone; Fingerprint; Reference point; Ridge analysis; Core-point detection; Camera
• ISSN: 1475-925X; 1475-925X
• DOI: 10.1186/s12938-015-0029-1

Background Fingerprint recognition systems essentially require core-point detection prior to fingerprint matching. The core-point is used as a reference point to align the fingerprint with a template database. When processing a larger fingerprint database, it is necessary to consider the core-point during feature extraction. Numerous core-point detection methods are available and have been reported in the literature. However, these methods are generally applied to scanner-based images. Hence, this paper attempts to explore the feasibility of applying a core-point detection method to a fingerprint image obtained using a camera phone. Method The proposed method utilizes a discrete wavelet transform to extract the ridge information from a color image. The performance of proposed method is evaluated in terms of accuracy and consistency. These two indicators are calculated automatically by comparing the method’s output with the defined core points. Results The proposed method is tested on two data sets, controlled and uncontrolled environment, collected from 13 different subjects. In the controlled environment, the proposed method achieved a detection rate 82.98%. In uncontrolled environment, the proposed method yield a detection rate of 78.21%. Conclusion The proposed method yields promising results in a collected-image database. Moreover, the proposed method outperformed compare to existing method.