The Recognition of New and Old Banknotes Based on SVM
Authors:
- Baiqing SunBSBaiqing Sun
- Jilu LiJLJilu Li
- 1 December 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2, 95-98
Abstract
On basis of reviewing the existed research results, and with digital image processing technology, this paper brings forward a rapid algorithm of detecting new and old banknotes, in which the Support Vector Machines (SVM) and Dynamic Time Warp (DTW) are combined. First, the banknotes gray histogram is aligned by DTW. Second, the aligned Gray Histogram is embedded in radial basis function (RBF) of SVM. Then, a mixed structure RBF/DTW is built. The simulation results show that the proposed structure is efficient to recognize the banknote levels of new and old.Keywords
This publication has 5 references indexed in Scilit:
- Employing optimized combinations of one-class classifiers for automated currency validationPattern Recognition, 2004
- Statistical pattern recognition: a reviewIEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
- Multicategory Classification by Support Vector MachinesComputational Optimization and Applications, 1999
- On the use of support vector machines for phonetic classification1999
- The Nature of Statistical Learning Theory1995