<![CDATA[The government has socialized the 3D method (Visible, Feeling, and Diterawang) to prevent people from accidentally transacting using counterfeit money. In addition to this method, counterfeit money scanners that use ultraviolet light are also commonly used. However, the use of methods or scanners must be accompanied by a human eye to determine whether the money is genuine or false. Along with the development of technology, the technique of making counterfeit money will also develop so that supporting techniques are needed that can be used to help identify counterfeit money automatically. In this study, the data used is only original and fake banknotes with a nominal value of Rp. 100,000.00. Image acquisition is done with the help of a digital camera and a light source from ultraviolet lights. Pre-processing in the form of pruning on the image of banknotes produces an image measuring 210 × 64 pixels. The feature extraction stage is done by changing the RGB image value to the HSV color space. The HSV value of the image is then used as an input vector on the LVQ network to determine the authenticity of the banknote image. The method test used for 100 real money images and 80 fake images yields the best performance when the learning level α (0) = 0,1 in epoch = 4 with an MSE value of 0,088889. Classification with the best accuracy occurs when the rate of learning α (0) = 0,1 in epoch = 1 with the level of truth classification is 87,2%.]]>
Copyrights © 2018
