Performance evaluation of feature selection methods for ANN based iris recognition

Iris recognition is receiving increasing attention as a means of personal recognition. Statistical methods, namely Single Value Decomposition (SVD), Principal Component Analysis (PCA) and Independent Component Analysis (ICA) are employed to extract the iris feature from a pattern named IrisPattern based on the iris image. These extracted patterns are classified by using a Feedforward Backpropagation Neural Network (BPNN) with different dimensions of features. From the experimental result, it is observed that ICA is the most appropriate feature extraction method for BPNN for the data sets used.