Gaussian Mixture Models for on-line signature verification
WBMA '03 Proceedings of the 2003 ACM SIGMM workshop on Biometrics methods and applications
A writer identification system for on-line whiteboard data
Pattern Recognition
Distributed Continuous Action Recognition Using a Hidden Markov Model in Body Sensor Networks
DCOSS '09 Proceedings of the 5th IEEE International Conference on Distributed Computing in Sensor Systems
Development of a Sigma-Lognormal representation for on-line signatures
Pattern Recognition
An efficient online signature verification scheme using dynamic programming of string matching
ICHIT'11 Proceedings of the 5th international conference on Convergence and hybrid information technology
Proceedings of the 2nd Conference on Wireless Health
A study on enhanced dynamic signature verification for the embedded system
BVAI'05 Proceedings of the First international conference on Brain, Vision, and Artificial Intelligence
Automatic Segmentation and Recognition in Body Sensor Networks Using a Hidden Markov Model
ACM Transactions on Embedded Computing Systems (TECS) - Special Section on CAPA'09, Special Section on WHS'09, and Special Section VCPSS' 09
| Hi-index | 0.00 |
In this paper, we propose a method of on-line signature verification system using HMM models in polar coordinate system. In the previous works, the signature verification was performed in terms of a standard x-y coordinate system. This coordinate system need to normalization ofsignature stroke size and angle variance always. To reduce normalization error and computing time, we proposed the polar coordinate system. Under the polar coordinate system, the feature sets are robust and independent of size and angle variation according to the user characteristics. The use of HMMs instead of a traditional Euclidian distance metric for the determination of the degree of match between a test signature and a reference signaturemodel brings about a significant improvement in performance as it permits the incorporation of heuristics. The experiments shows that the proposed method gives lower equal error rate 2.2% from 50,000 testing case.