Enhancement of the Iris-Texture by Removal of Specular Reflections for an accurate Iris Segmentation
Imtiyaz Ahmed B K, Ghouse Ahmed, Abdul Saleem, Shabeer Ahmed · 2020 Second International Conference on Inventive Research in Computing Applications (ICIRCA) · 2020
In the area of authentication using biometrics, Iris match is considered as one of the most efficient method. One of the challenges to be addressed in such method is Specular Reflection that is because of reflective nature of human eye structure. As this biometric system needs processing of iris images, they have to deal with noise due to this inherent reflection. To achieve expected authenticity results, it would be necessary to remove this specular reflection to enhance the iris texture in the captured images. A method to enhance the iris structure by removing any reflection by using Top hat filtering has been implemented. A related graphical user interface (GUI) has also been developed for the proposed work. Simulation results in terms of mean square error (MSE) in comparison with other two methods namely, four-pixel connectivity and bilinear interpolation are presented. It was observed that the proposed method produces best results compare to other two methods.