Feature Selection for Iris Recognition with AdaBoost
Kan-Ru Chen, Chia-Te Chou, Sheng-Wen Shih, Wen‐Shiung Chen, Duan-Yu Chen · 2007
In this paper, we proposed a method for selecting edge-type features for iris recognition. The AdaBoost algorithm is used to select a filter bank from a pile of filter candidates. The decisions of the weak classifiers associated with the filter bank are linearly combined to form a strong classifier. Real experiments have been conducted to assess the performance of the designed strong classifier. The results showed that the boosting algorithm can effectively improve the recognition accuracy at the cost of slightly increase the computation time.