Evaluation of Statistical Feature Encoding Techniques on Iris Images

Feature selection, often used as a pre-processing step to machine learning, is designed to reduce dimensionality, eliminate irrelevant data and improve accuracy. Iris basis is our first attempt to reduce the dimensionality of the problem while focusing only on parts of the scene that effectively identify the individual. Independent component analysis (ICA) is to extract iris feature to recognize iris pattern. Principal component analysis (PCA) is a dimension-reduction tool that can be used to reduce a large set of variables to a small set that still contains most of the information in the large set. Image quality is very important in biometric authentication techniques. We have assessed the collision of various factors on performance of ICA and PCA as well as evaluated which factors can be plausibly compensated on iris patterns.