An Iris preprocessing and segmentation procedures for Iris Recognition: the darkest crescent algorithm based on the RGB space
Faiza Walha, Houda Khmila, Imène Khanfir Kallel · 2022 8th International Conference on Control, Decision and Information Technologies (CoDIT) · 2022
In this study, a new method of Image preprocessing and segmentation based on the RGB space is presented. The method consists of preprocessing the images by cleaning and removing unnecessary areas then applying a double search for the darkest areas to find the iris and the pupil. For experimental results, the study is implemented on Iris@PBMLT dataset which is a new private IRIS dataset. The results are compared with those that are obtained by the classical integro-differential operator method. The obtained results showed that the proposed method has acceptable accuracy.