An Implementation-Oriented Iris Recognition Algorithm Using Phase-Based Image Matching

摘要

This paper presents a phase-based iris recognition algorithm which is specially designed for system implementation. In order to reduce the size of registration iris data and to prevent the visibility of iris images, we introduce the idea of 2D Fourier phase code (2D FPC) for representing iris information. 2D FPC corresponds to the quantized version of phase-spectrum of an iris image, which is essential for phase-based iris recognition. By the use of 2D FPC, while keeping a sufficient level of performance, the size of iris data can be reduced to below one-quarter compared to using iris image directly as the registration data. 2D FPC is particularly useful for implementing compact iris recognition devices using state-of-the-art DSP (digital signal processing) technology