COUNTERFIET PAPER BANKNOTE IDENTIFICATION BASED ON COLOR AND TEXTURE

Abstract

Currency identification is the application of systematic methods to determine authenticity of questioned currency. However, identification analysis is a difficult task requiring specially trained examiners; the most important challenge is automating the analysis process reducing human labor and time. Color and texture feature are used for the classification of an image. This thesis present the techniques used for the extraction of feature, identification and classification of counterfeit and genuine bank note. It presented simple method of identification of counterfeit paper banknotes, which automatically using image processing techniques. Color and texture feature of a currency is used for identification. Color descriptor skew, mean and standard deviation is calculated from samples which are checked against the parameter that are previously defined. Texture parameter entropy and correlation are calculated from different set of database image. Matching score below the threshold, input currency image is classified as fake note. Otherwise the currency is genuine. Image processing approach is very useful for classification of different types of currency.