Pupil Lovalization And Tracking For Video-Based Iris Biometrics
Abstract
In this paper, we are interested in iris biometric applications. More precisely, our contribution consists in designing both a pupil detection and a tracking procedure from video sequences acquired by low-cost webcams. The novelty of our approach relies on the fact that it is operational even with a minimal user cooperation and, under bad illuminations and acquisition conditions. A robust classification algorithm is designed to detect the pupil. Moreover, a pupil tracker based on the extended Kalman filter is applied in order to reduce the processing time. Experimental results are performed in order to evaluate the performances of the proposed detection and tracking system
Related Papers
SCISSOR: Mitigating Semantic Bias through Cluster-Aware Siamese Networks for Robust Classification2025-06-17Quantum-Informed Contrastive Learning with Dynamic Mixup Augmentation for Class-Imbalanced Expert Systems2025-06-16CEM-FBGTinyDet: Context-Enhanced Foreground Balance with Gradient Tuning for tiny Objects2025-06-11Inherently Faithful Attention Maps for Vision Transformers2025-06-10Identifying and Understanding Cross-Class Features in Adversarial Training2025-06-05Uncertainty-Aware Genomic Classification of Alzheimer's Disease: A Transformer-Based Ensemble Approach with Monte Carlo Dropout2025-05-31Acoustic Classification of Maritime Vessels using Learnable Filterbanks2025-05-29CNN-LSTM Hybrid Model for AI-Driven Prediction of COVID-19 Severity from Spike Sequences and Clinical Data2025-05-29