Recent studies have concentrated on employing various machine learning approaches to determine the genuineness of banknotes. The task of accurately distinguishing authentic notes from counterfeit ones is difficult. In the current study, two distinct classes of banknotes were used to train a deep neural network (DNN) and random forest to recognize the legitimacy of the notes. Both approaches’ experimental outcomes have been contrasted with one another. To compare the algorithms’ performances, Performance metrics such as accuracy, precision, recall, and F1-score-measure have been employed. The experimental findings show that DNN significantly outperforms random forest in detecting banknote fraud.