Research on Methods of Features Extraction for Identifying False RMB Ultraviolet Images

摘要

In order to solve the problem of feature selection for the identification of RMB Ultraviolet Images, a method of texture feature extraction image based on Gabor transformation is presented. As we all know, the texture intensity of true currency in each direction is more obvious then counterfeit notes and the difference is more prominent. In this paper, the Gabor transform is carried out in 8 directions. In other words, 8 texture features from RMB are extracted. In order to effectively carry out the false, we gave a full analysis on the image we did. Algorithm proposed in this paper is verified on feature selection of RMB. Furthermore, experimental results show that using this method can intensity currency accurately. This technology has certain utility value and it will be widely used in daily life.