Despite the rapidly increasing use of credit cards and other electronic forms of payment, cash is still widely used for everyday transactions due to its convenience, perceived security and anonymity. However, the visually impaired might have a hard time telling each paper bill apart, since, for example, all dollar bills have the exact same size and, in general, currency bills around the world are not distinguishable by any tactile markings. We propose the use of a broadly available tool, the camera of a smart-phone, and an adaptation of the SIFT algorithm to recognize partial and even distorted images of paper bills. Our algorithm improves memory efficiency and the speed of SIFT key-point classification by using a k-means clustering approach. Our results show that our system can be used in real-world scenarios to recognize unknown bills with a high accuracy.