Toward Accurate and Reliable Iris Segmentation Using Uncertainty Learning

As an upstream task of iris recognition, iris segmentation plays a vital role in multiple subsequent tasks, including localization and matching. A slight bias in iris segmentation often results in obvious performance degradation of the iris recognition system. In the paper, we propose an Iris U-transformer (IrisUsformer) for accurate and reliable iris segmentation. For better accuracy, we elaborately design IrisUsformer by adopting position-sensitive operation and re-packaging transformer block to raise the spatial perception ability of the model. For better reliability, IrisUsformer utilizes an auxiliary head to distinguishes the high- and low-uncertainty regions of segmentation predictions and then adopts a weighting scheme to guide model optimization. Experimental results on three publicly available databases demonstrate that IrisUsformer achieves better segmentation accuracy using 35% MACs of the SOTA IrisParseNet. More importantly, our method estimates the uncertainty map corresponding to the segmentation prediction for subsequent processing in iris recognition systems.