Combining fractional-order edge detection and chaos synchronisation classifier for fingerprint identification

• Published in: IET image processing Vol. 8; no. 6; pp. 354 - 362
• Main Authors: Chen, Jian-Liung, Huang, Cong-Hui, Du, Yi-Chun, Lin, Chia-Hung
• Format: Journal Article
• Language: English
• Published: Stevenage The Institution of Engineering and Technology 01.06.2014; Institution of Engineering and Technology; The Institution of Engineering & Technology
• Subjects: Applied sciences; chaos synchronisation classifier; Chaos theory; chaotic communication; Classifiers; Detection, estimation, filtering, equalization, prediction; dynamic errors; Dynamics; Edge detection; Exact sciences and technology; feature extraction; fingerprint identification; Fingerprints; FOED; fractional‐order edge detection; grey‐scale values; heterogeneous fingerprints; Image enhancement; Image processing; image sampling; Information, signal and communications theory; integral‐order method; maximum likelihood estimation; maximum‐likelihood method; multiple CS‐based detectors; optical fingerprint reader; pattern classification; pattern identification; Pattern recognition; Readers; ridge structures; Sampling, quantization; Signal and communications theory; Signal processing; Signal, noise; stable sampling window; synchronisation; Telecommunications and information theory; thumbprint images; valley structures; fingerprint identification; integral-order method; stable sampling window; pattern classification; FOED; optical fingerprint reader; grey-scale values; maximum-likelihood method; chaos synchronisation classifier; maximum likelihood estimation; multiple CS-based detectors; synchronisation; fractional-order edge detection; chaotic communication; image sampling; heterogeneous fingerprints; image enhancement; feature extraction; pattern identification; ridge structures; dynamic errors; edge detection; thumbprint images; valley structures; Biometrics; Chaos; Automatic classification; Image recognition; Image processing; Processing time; Information extraction; Shape detection; Sampling; Fingerprint; Target tracking; Measurement sensor; Singular point; Pattern recognition; Synchronization; Signal classification; Integral method; Image quality; Gray scale; Fingerprint identification; Signal processing; Feature extraction; Optical reader; Maximum likelihood; Automatic recognition; Edge detection; Image enhancement
• ISSN: 1751-9659; 1751-9667
• DOI: 10.1049/iet-ipr.2012.0660

This study proposes the combination of fractional-order edge detection (FOED) and a chaos synchronisation classifier for fingerprint identification. Fingerprints have various morphologies and exhibit singular points, which result in fingerprint individuality. Thumbprint images are captured from subjects using an optical fingerprint reader. The identification procedure consists of three stages: image enhancement, feature extraction and pattern identification. The adjustment of grey-scale values is used to enhance the contrast of the image. In order to overcome the limitations of the integral-order method, FOED is used to improve the clarity of the ridge and valley structures in fingerprint images. Using a reference point, it provides a stable sampling window for fingerprint extraction. Multiple CS-based detectors are used to track the differences as dynamic errors between heterogeneous fingerprints, on a one-to-one basis. The maximum-likelihood method performs a comparison of these different dynamic errors to identify individuals. Using 30 laboratory subjects, the proposed hybrid methods have a faster processing time and provide more accurate fingerprint identification.