Title :
Selecting a Reference High Resolution for Fingerprint Recognition Using Minutiae and Pores
Author :
Zhang, David ; Liu, Feng ; Zhao, Qijun ; Lu, Guangming ; Luo, Nan
Author_Institution :
Dept. of Comput., Hong Kong Polytech. Univ., Kowloon, China
fDate :
3/1/2011 12:00:00 AM
Abstract :
High-resolution automated fingerprint recognition systems (AFRSs) offer higher security because they are able to make use of level-3 features, such as pores, that are not available in lower resolution ( <; 500-dpi) images. One of the main parameters affecting the quality of a digital fingerprint image and issues such as cost, interoperability, and performance of an AFRS is the choice of image resolution. In this paper, we identify the optimal resolution for an AFRS using the two most representative fingerprint features: minutiae and pores. We first designed a multiresolution fingerprint acquisition device to collect fingerprint images at multiple resolutions and captured fingerprints at various resolutions but at a fixed image size. We then carried out a theoretical analysis to identify the minimum required resolution for fingerprint recognition using minutiae and pores. After experiments on our collected fingerprint images and applying three requirements for the proportions of minutiae and pores that must be retained in a fingerprint image, we recommend a reference resolution of 800 dpi. Subsequent tests have further confirmed the proposed reference resolution.
Keywords :
feature extraction; fingerprint identification; image resolution; open systems; AFRS; digital fingerprint image quality; high-resolution automated fingerprint recognition system; image resolution; interoperability; level-3 feature; multiresolution fingerprint acquisition device; Fingerprint recognition accuracy; high-resolution automated fingerprint recognition systems (AFRSs); minutiae; pores; selecting resolution criteria;
Journal_Title :
Instrumentation and Measurement, IEEE Transactions on
DOI :
10.1109/TIM.2010.2062610