Iris recognition based on location of key points

This paper proposes a new iris recognition method based on the location of key points. When preprocessed, the annular iris is normalized into a rectangular image. Multi-channel 2-D Gabor filters are used to capture the feature of iris texture. In each filtered sub-image, points that can represent the local texture feature most effectively in each channel are extracted. The barycenter of these points is the so-called key point. For a given iris, the location of key points is recorded as its feature vector. The iris feature matching is based on the Euclidean distance between the corresponding key points. Experimental results illustrate the effectiveness of this method.