Segmentation Enhancement for Iris Recognition Using Unit Gradient Vectors
Limhourlaurent Meam, Suradej Duangpummet, Waree Kongprawechnon · 2023
Iris recognition is a prominent technique in biometric identification due to its robustness and distinctive features. However, challenges in iris segmentation, including limitation of sensors, occlusion, and inconsistent lighting, significantly impact the accuracy and efficiency of the recognition. Therefore, this paper proposes improving iris segmentation by incorporating a technique known as Unit Gradient Vectors (UGVs) into the traditional preprocessing stage. The UGVs at each pixel provide edge direction, enabling the accentuation of areas with significant directional changes. Thus, a more robust and correct edge detection from inconsistent lighting images can be obtained. An experiment was conducted using the CASIA-Iris-Lamp, which has challenging lighting conditions. Results suggest that the proposed method achieves superior performance compared to the existing method. The accuracy, false acceptance and rejection rates were improved by 2.04%, 10.00%, and 7.33%, respectively. Specifically, the equal error rate was dramatically improved by 37.78%. These promising results can contribute to advancing security and efficiency in iris recognition systems.