US Patent 10380418 Iris recognition based on three-dimensional signatures

A first image set that includes a plurality of 2D images of an eye of a user is collected. Two or more sets of corresponding pixels are generated from the plurality of 2D images. One or more 3D features of an iris of the user is extracted based on repeatable differences in reflectance among each set of corresponding pixels. A test set of 3D features for a submitted eye is generated based on a second image set, the second image set including a plurality of 2D images of the submitted eye. Based on a comparison of the one or more extracted 3D features and the test set of 3D features, an indication is made as to whether the second image set is representative of the user's eye.