This article proposes a new approach for iris recognition.
The authors contend that traditional iris recognition algorithms work well for frontal iris images; however, when the angle of an eye changes with respect to the camera lens (“off-angle”) the size, shape, and detail of iris patterns change as well and cannot be matched using traditional methods. The new method proposed does not require polar transformation, affine transformation, or highly accurate segmentation to perform iris recognition. The authors research show positive results in that the proposed method works and is successful for both frontal look images and off-angle images.
810 Seventh Street NW, Washington, DC 20531, United States