Comparisons of anterior segment biometry between Chinese and Caucasians using anterior segment optical coherence tomography.

PURPOSE: To compare anterior segment parameters between eyes of Chinese and Caucasians using anterior segment optical coherence tomography and to evaluate the association between these parameters and anterior chamber angle width between the two ethnic groups. METHODS: 60 Chinese and 60 Caucasians, 30 with open angles and 30 with narrow angles (defined as Shaffer grade < or =2 in > or =3 quadrants during dark room gonioscopy) in each group, were consecutively enrolled. One eye of each subject was randomly selected for imaging in a completely darkened room. Measurements, including anterior chamber depth (ACD), scleral spur-to-scleral spur distance (anterior chamber width (ACW)), anterior chamber angle width, iris convexity and iris thickness, were compared between the groups. The associations between angle opening distance and biometric measurements were evaluated with univariate and multivariate regression analyses. RESULTS: There were no differences in age, axial length, anterior chamber angle measurements, pupil diameter and iris convexity between Chinese and Caucasians in both open-angle and narrow-angle groups. However, the ACD and ACW were smaller and the iris was thicker in Chinese. In the multivariate analysis, the ACD was the most influential biometric parameter for angle opening distance in both Chinese and Caucasians. After adjusting the effects of axial length, age and sex, ACD and ACW were significantly smaller in Chinese. CONCLUSIONS: Chinese eyes had smaller ACD, smaller ACW and greater iris thickness than Caucasians. ACD was the most influential parameter in determining the angle width in both ethnic groups.

Face Recognition in the Scrambled Domain via Salience-Aware Ensembles of Many Kernels

With the rapid development of internet-of-things (IoT), face scrambling has been proposed for privacy protection during IoT-targeted image/video distribution. Consequently in these IoT applications, biometric verification needs to be carried out in the scrambled domain, presenting significant challenges in face recognition. Since face models become chaotic signals after scrambling/encryption, a typical solution is to utilize traditional data-driven face recognition algorithms. While chaotic pattern recognition is still a challenging task, in this paper we propose a new ensemble approach – Many-Kernel Random Discriminant Analysis (MK-RDA) to discover discriminative patterns from chaotic signals. We also incorporate a salience-aware strategy into the proposed ensemble method to handle chaotic facial patterns in the scrambled domain, where random selections of features are made on semantic components via salience modelling. In our experiments, the proposed MK-RDA was tested rigorously on three human face datasets: the ORL face dataset, the PIE face dataset and the PUBFIG wild face dataset. The experimental results successfully demonstrate that the proposed scheme can effectively handle chaotic signals and significantly improve the recognition accuracy, making our method a promising candidate for secure biometric verification in emerging IoT applications.