Penggunaan Algoritma Color-Based Filtering Sebagai Pendeteksi Nominal Pada Uang Kertas

As a legal tender, banknotes have several drawbacks in terms of their use. for people with visual impairments they have, it is very likely that they will be confused, misplaced, and even some ignorant people take advantage of their weaknesses in using the money. Based on this problem, in this paper we propose to make a banknote detector with sound output. The method used for the nominal identification process is the color based filtering algorithm using the Pixy2 CMUcam5 camera and Arduino as the data processor. Tests were carried out on 1000, 2000, 5000, 10000, 20000 and 50000 banknotes. From the test results, it was found that the success rate was 90% with the optimal distance for the camera scan process being 7 to 10 cm in an average time of about 5.7 seconds.