Efficient Iris Recognition Using Adaptive Quotient Thresholding

Abstract

This paper presents an intensity-based iris recognition system. The system exploits local intensity changes of the visible iris textures such as crypts and naevi. The textures are extracted using local histogram equalization and the proposed 'quotient thresholding' technique. The quotient thresholding partitions iris images in a database such that a ratio between foreground and background of each image is retained. By fixing this ratio, variations of illumination across iris images are compensated, resulting in informative and distinctive blob-like iris textures. An agreement of the two extracted textures is measured by finding spatial correspondences between the textures. The proposed system yields the 0.22 %EER and 100%CRR. The experimental results indicate encouraging and effective iris recognition system, especially when it is used in identification mode. The system is very robust to changes in decision ratio.

Two Approaches Based on Genetic Algorithm to Generate Short Iris Codes

“…Some authors directly used the structural iris patterns to generate the iris feature vectors. Segmenting the blobs of interest (BOI) by zero-crossing wavelet [14], local h istogram equalization and a quotient thresholding [15], extract ing iris speckles [16] and shape analysis techniques for near infrared (NIR) and visible light (VL) images and fusing them [17], are the handful of structural iris representation. The implemented systems by all of the above methods have shown the encouraging experimental results.…”

Iris Recognition: Robust Processing, Synthesis, Performance Evaluation and Applications

“…Extracted features were quantized to two levels yielding binary codes. Techniques other than filter-based approaches can also be used to generate binary iris codes, for example, local histogram equalization and a quotient thresholding [18], [19], [20].…”

“…These include modified log-Gabor filters [9], non-local comparisons [10], dyadic wavelet transforms [13], derivative-of-Gaussians [12], Laplacian-of-Gaussian filters [12], [11], discrete cosine transforms [14], wavelet packets [15], [16], [17], and quotient thresholding techniques [18]. Our method falls into the local ordinal measures category [10] and is similar to the thresholding approach of [18]. This method used local histogram equalization to compensate non-uniform illumination, and adaptive quotient thresholding to keep the ratio between the foreground and background constant.…”