Abstract: The inner and the outer edges of an iris image in the Cartesian coordinates are transformed into quas-i lines in the polar coordinates,with the origin of the coordinate transformation being roughly determined through the projection of the grey-level pixel values.Then the inner circular edge and its center are obtained in an iterative fashion with the aid of some specially chosen auxiliary points. After contrast enhancement and smoothing preprocessing,the outer edge is located with an improved Canny operator and a curve fitting technique.Experimental results show that the performance of the proposed algorithm is superior over the traditional method based on line detection,and the average time for iris location is0.1second,a satisfactory rate for rea-l time iris recognition.