The Research On The Method Of Iris Location

Iris recognition is a new technology of biometrics, which has been paid attention to by more and more scientists because of its unique feature and applying angle. Iris alone can serve as a biometrics in some occasions, and it can also be combined with other biometrics to act as multimode biometrics. The unique characters of iris make it possible to enrich biometrics technology and to make up the disadvantages of other biometrics.Iris location is one of the important processes in Iris recognition. There are many classical iris location methods, such as Daugman's method and the method of edge-detection operator combing with Hough transform. Some could locate the iris accurately, but they would spend much time. Others may need a short time, but have low accuracy of location because of the complicated image. In fact, the disturbances of the eyelid and the eyelashes are very hard when the eye opens in the normal state. When the eye doesn't open largely enough, we may mistake the boundary of the eyelid and the Iris for that of the iris and the sclera. Meanwhile, the upper eyelashes's disturbance may make us mistake the point on the upper eyelashes for the left-hand or the right-hand point of the iris's outer boundary. So the Iris's outer boundary that we have deduced isn't the actual outer boundary, which will increase the wrong probability. An iris location method based on the structure of human eyes is proposed in the paper. It could locate the iris boundary accurately and rapidly.Large numbers of experiment have been done in three iris databases. Especially, one of them which include 1200 complicated images is heavily located. The proposed method's accuracy of location reaches 99.08%, average time of location reaches 0.8sec. It could well locate the iris image, which has large numbers of disturbances in it. So the proposed algorithm is an effective and feasible method.