Redundancy Reduction of Features by Independent Component Analysis

We propose the method to reduce the redundancy be tween features extracted from images by the infomax a1gorithm. The method consisted of the addition of noise and the feature se1ection based on kurtosis. The addition of noise to images contributes to the reduction of redundancy and make it easy to set a criterion of feature se1ection. We adopt ETL-4 database sets of hand-written Japanese Hiragana characters ， and performed the recognition experiment with the extracted features. Results of the recognition experiment showed that it was possib1e to reduce the redundancy of features by the proposed method.