Dynamics features Extraction for on-Line Signature verification
CONIELECOMP '04 Proceedings of the 14th International Conference on Electronics, Communications and Computers
On-Line Signature Verification with Hidden Markov Models
ICPR '98 Proceedings of the 14th International Conference on Pattern Recognition-Volume 2 - Volume 2
ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
| Hi-index | 0.00 |
Signature verification and recognition can be divided into online and offline, depending on the sensing modality. In an online method, the dynamic information signature features such as direction, slant, baseline, pressure, speed and numbers of pen ups and downs can be captured. Method of extracting features signature depends on the requirement features to be extracted. Thus this paper discusses the construction of an algorithm to extract the baseline from signature. The Signature Extraction Features System (SEFS) uses the proposed algorithm and provides a set of tools that allow the systems to extract baseline features in signature automatically for analysis purposes. Signatures are taken from twenty randomly selected individuals with different background. These individuals will have to sign their signatures on the tablet and the SEFS system will gather and store the raw data. SEFS created the image of each signature and these images were used as samples for a developed questionnaire to be given to human expert for evaluation. These questionnaires are all about identifying the baseline of the signatures. Both results from SEFS and the questionnaire are compared, and it shows that the algorithm is 90% accurate. It can be concluded that the algorithm proposed are acceptable to represent extraction of signature features based on baseline.