Eliza Yingzi Du

A New Human Identification Method: Sclera Recognition

The blood vessel structure of the sclera is unique to each person, and it can be remotely obtained nonintrusively in the visible wavelengths. Therefore, it is well suited for human identification (ID). In this paper, we propose a new concept for human ID: sclera recognition. This is a challenging research problem because images of sclera vessel patterns are often defocused and/or saturated and, most importantly, the vessel structure in the sclera is multilayered and has complex nonlinear deformations. This paper has several contributions. First, we proposed the new approach for human ID: sclera recognition. Second, we developed a new method for sclera segmentation which works for both color and grayscale images. Third, we designed a Gabor wavelet-based sclera pattern enhancement method to emphasize and binarize the sclera vessel patterns. Finally, we proposed a line-descriptor-based feature extraction, registration, and matching method that is illumination, scale, orientation, and deformation invariant and can mitigate the multilayered deformation effects and tolerate segmentation error. The experimental results show that sclera recognition is a promising new biometrics for positive human ID.

Multi-angle sclera recognition system

Sclera patterns can be used for human classification and identification; however image quality can significantly affect the recognition accuracy. In this paper, we studied and analyzed four multi-angle sclera recognition fusion methods. The experimental results show that these proposed multi-angle sclera recognition systems can achieve better performance in general. In addition, it shows that it is important to take system application needs into account when selecting the fusion method.

A multi-stage approach for non-cooperative iris recognition

In this paper, we proposed a robust multi-stage feature extraction and matching approach for non-cooperative iris recognition. To speed up the process, we used the SURF concept that treats the image as discrete and digital matrices. Unlike SURF method, we applied Gabor wavelets for feature extraction. More importantly, we designed a multi-scale feature point extraction method and a multi-stage matching scheme to improve the recognition accuracy and speed. The experimental results show that the proposed method can not only improve the process speed but also achieve better recognition accuracy.

Automatic categorization-based multi-stage pedestrian detection

Pedestrian safety has become an important issue for automobile design. Although a lot of research has been done or is ongoing in developing in-car camera-based pedestrian protection systems, robust and reliable in-car camera based pedestrian analysis is still very challenging, especially for real-time systems or large scale dataset analysis. In this paper, we propose a new pedestrian detection and analysis system based on automatic categorization. A category-based multi-stage pedestrian detection and data analysis approach is developed to efficiently process the extremely large scale driving data collected in this research. The experimental results on part of the collecting dataset show that the proposed method is promising.

Quality fusion based multimodal eye recognition

Multimodal eye recognition can improve the biometric systems recognition accuracy by combining iris and sclera recognition. However, poor quality images can significantly affect the system performance. In this paper, we proposed a quality fusion based multimodal eye recognition. Our quality measure evaluated the entire eye image quality, iris area quality, and sclera area quality. The experimental results show that our overall iris and sclera quality scores are highly correlated to recognition accuracy, and our quality fusion based eye recognition can improve and predict the performance of eye recognition systems.

Pilot study on pedestrian step frequency in naturalistic driving environment

Investigation of pedestrian step frequency is essential for analyzing walking gaits and pedestrian behaviors. However, most research about step frequency is performed in labs or manually controlled experimental environment, which greatly limits the utilization of the results to analyze and/or predict real pedestrian behaviors. This study investigates the step frequencies of pedestrian in naturalistic driving environment. The mean step frequency values and distribution are studied in all cases and separately for road crossing cases only. Furthermore, comparisons of pedestrian step frequencies are made considering three different impact factors. The results have shown that in real world, people tend to use higher step frequencies when crossing the road, especially when the vehicle is moving towards the pedestrian or when the pedestrians are crossing without right-of-way.

Secure and privacy-preserving biometrics based active authentication

User authentication is critical in preventing system breaches. Existing authentication approaches usually do a onetime log-in authentication, but rarely incorporate mechanisms to differentiate the initial log-in user and the user who is currently taking control of the system, which may cause post-authentication breaches. In this paper, we study user authentication for both login session and post-authentication session and propose a biometrics based active authentication approach. Moreover, concerning the usage of biometrics, the system is biometrics-secure and privacy-preserving. Security analysis and experimental results prove that the proposed approach is secure, resilient to various attacks and effective.

New region feature descriptor-based image registration method

Successful content-based image registration relies on the accurate identification of corresponding features across images. Geometric and photometric transformations between images may hinder an algorithms ability to precisely match features. In this work, we propose a novel region descriptor detection and matching algorithm for use with image registration. The detection process utilizes invariant feature points, as well as their spatial relationships and textural characteristics to create a connected graph whose structure represents an invariant region descriptor. With such a framework, feature matching can be accomplished by graph matching with a defined similarity metric. Subsequent image registration steps are outlined that employ the invariant region descriptors. The results provide strong evidence of the region descriptors effectiveness in applications involving image registration. Several scenarios are presented including the registration of general objects, aerial photography, as well as scenes before and after a disaster.

A novel graph-based invariant region descriptor for image matching

Identification of invariant image descriptors is an integral task for many computer vision applications such as image registration, object recognition, and object tracking. The detected features should be invariant to geometric transformations such as rotation and translation, as well photometric variations due to differing lighting conditions. In this work, we propose a unique and effective region descriptor that couples invariant features and texture information. The descriptor relies on spatial relationships of invariant SURF features to create a graph-based descriptor for image matching. Additionally, a novel method is proposed for matching region descriptors through the definition of an efficient similarity measure that couples invariant features and their spatial relationships. Several examples are presented to illustrate the effectiveness of the proposed region descriptor while the results of the proposed approach outperform SURF feature point matching.

Key Incorporation Scheme for Cancelable Biometrics

Biometrics is becoming an important method for human identification. However, once a biometric pattern is stolen, the user will quickly run out of alternatives and all the applications where the associated biometric pattern is used become insecure. Cancelable biometrics is a solution. However, traditional cancelable biometric methods treat the transformation process and feature extraction process independently. As a result, this kind of cancelable biometric approach would reduce the recognition accuracy. In this paper, we first analyzed the limitations of traditional cancelable biometric methods, and proposed the Key Incorporation Scheme for Cancelable Biometrics approach that could increase the recognition accuracy while achieving “cancelability”. Then we designed the Gabor Descriptor based Cancelable Iris Recognition method that is a practical implementation of the proposed Key Incorporation Scheme. The experimental results demonstrate that our proposed method can significantly improve the iris recognition accuracy while achieving “cancelability”.