In this paper, an efficient classification algorithm using principal components analysis for enhancing the performance of the currently used bank note classification system is proposed. This method first transforms four feature vectors from sensing de...
In this paper, an efficient classification algorithm using principal components analysis for enhancing the performance of the currently used bank note classification system is proposed. This method first transforms four feature vectors from sensing devices into principal components analysis, then maps the feature vectors to eigenvectors corresponding to maxium among eigenvalues. A cash is classified as new if it is greater than the threshold set by user, as old if it is not. The experimental result shows that the proposed system enhance the performance of the current system above 1%.