Multi-stage Visible Wavelength and Near Infrared Iris Segmentation Framework

摘要: This paper presents a multi-stage iris segmentation framework for the localization of pupillary and limbic boundaries human eyes. Instead applying time-consuming exhaustive search approaches, like traditional circular Hough Transform or Daugman's integrodifferential operator, an iterative approach is used. By decoupling coarse center detection fine boundary localization, faster processing modular design can be achieved. alleviates more sophisticated quality control feedback during process. avoiding database-specific optimizations, this work aims at supporting different sensors light spectra, i.e. Visible Wavelength Near Infrared, without parameter tuning. The system evaluated by using multiple open databases it compared to existing classical approaches.