Scale Invariant Gabor Descriptor-Based Noncooperative Iris Recognition
- PDF / 3,410,313 Bytes
- 13 Pages / 600.05 x 792 pts Page_size
- 97 Downloads / 231 Views
Research Article Scale Invariant Gabor Descriptor-Based Noncooperative Iris Recognition Yingzi Du, Craig Belcher, and Zhi Zhou Department of Electrical and Computer Engineering, Indiana University-Purdue University Indianapolis, Indianapolis, IN 46202, USA Correspondence should be addressed to Yingzi Du, [email protected] Received 1 January 2010; Accepted 18 March 2010 Academic Editor: Robert W. Ives Copyright © 2010 Yingzi Du et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. A new noncooperative iris recognition method is proposed. In this method, the iris features are extracted using a Gabor descriptor. The feature extraction and comparison are scale, deformation, rotation, and contrast-invariant. It works with off-angle and lowresolution iris images. The Gabor wavelet is incorporated with scale-invariant feature transformation (SIFT) for feature extraction to better extract the iris features. Both the phase and magnitude of the Gabor wavelet outputs were used in a novel way for local feature point description. Two feature region maps were designed to locally and globally register the feature points and each subregion in the map is locally adjusted to the dilation/contraction/deformation. We also developed a video-based noncooperative iris recognition system by integrating video-based non-cooperative segmentation, segmentation evaluation, and score fusion units. The proposed method shows good performance for frontal and off-angle iris matching. Video-based recognition methods can improve non-cooperative iris recognition accuracy.
1. Introduction Performing noncooperative iris recognition is important for a number of tasks, such as video surveillance and watchlist monitoring (identifying most wanted criminals/suspects) [1–4]. In addition, noncooperative iris recognition systems can provide added convenience for cooperative users for identification [5]. However, it is challenging to design an iris recognition system that can work in a noncooperative situation, where the image quality may be low and the eye may be deformed, due to a nonfrontal gaze. In recent years, several methods have been developed for iris recognition [2, 3]. Most of these methods are designed for frontal and high-quality iris images. Among them, Daugman’s approach has been most widely used in the commercialized iris recognition systems [6–9]. This method transforms the segmented iris image into log-polar coordinates, extracts the iris features using a 2D Gabor wavelet, and encodes the phase information into a binary iris code [7, 9]. Hamming distance is used to match two iris codes [7]. Daugman’s method has been tested and evaluated using large databases, such as the United Arab Emirates (UAE)
database with over 600,000 iris images with over 200 billion comparisons [8]. Chen et al. proposed using Daugman’s 2D Gabor filter with quality measure enhancement to improve the recogniti
Data Loading...
