Signature Verification Using Static and Dynamic Features
A signature verification algorithm based on static and dynamic features of online signature data is presented. Texture and topological features are the static features of a signature image whereas the digital tablet captures in real-time the pressure values, breakpoints, and the time taken to create a signature. 1D – log Gabor wavelet and Euler numbers are used to analyze the textural and topological features of the signature respectively. A multi-classifier decision algorithm combines the results obtained from three feature sets to attain an accuracy of 98.18%.
Unable to display preview. Download preview PDF.
- 1.Bigun, J., du Buf, J.M.: N-folded symmetries by complex moments in Gabor space and their applications to unsupervised texture segmentation. IEEE Transactions on PAMI 16(1), 80–87 (1994)Google Scholar
- 2.Brault, J.-J., Plamondon, R.: Segmenting Handwritten Signatures at Their Perceptually Important Points. IEEE Transactions on PAMI 15(9), 953–957 (1993)Google Scholar
- 3.Daugman, J.: Recognizing Persons by their Iris Patterns. In: Jain, A., Bolle, R., Pankati, S. (eds.) Biometric: Personal Identification in Networked Society, pp. 103–121. Kluwer, Dordrecht (1998)Google Scholar
- 4.Gonzalez, W.: Digital Image Processing, 2nd edn. Pearson Education, LondonGoogle Scholar
- 5.Josef, K., Mohamad, H., Duin Robert, W.P., Jiri, M.: On combining classifiers. IEEE Transactions on PAMI 20(3), 226–239 (1998)Google Scholar
- 6.Rubner, Y., Tomasi, C.: Coalescing Texture Descriptors. In: Proceedings of the ARPA Image Understanding Workshop (1996)Google Scholar