Offline Signature Recognition and Verification using Neural Network
- Dhananjay Rakshe
- D. B. Kshirsagar
Feature Extraction, Neural Network, Back Propagation algorithm, Signature Verification.
Computers have become common and are used in almost every field including financial transactions, so it’s necessary to provide additional security measures. Considering consumers’ expectations, these security measures must be reliable, cheap and un-intrusive to the authorized person. This technique of signature recognition has advantage over the other biometric techniques like voice, iris, fingerprint etc. as it is mostly used for daily routine procedures like document analysis, banking operations, access control, electronic funds transfer etc. Most importantly, people are less likely to object it because its easy. The technique described in this paper uses neural network which enables the user to recognize whether a signature is original or a fraud. Scanned images are introduced into the computer, their quality is modified with the help of image enhancement and noise reduction techniques, specific features are extracted and neural network is trained. The different stages of the process involve image preprocessing followed by feature extraction and pattern recognition through neural networks. This method will be more efficient and provide more accurate results than the existing techniques.
Dhananjay Rakshe, D. B. Kshirsagar. "Offline Signature Recognition and Verification using Neural Network".INTERNATIONAL JOURNAL OF ENGINEERING DEVELOPMENT AND RESEARCH ISSN:2321-9939, Vol.3, Issue 3, pp., URL :https://rjwave.org/ijedr/papers/IJEDR1503016.pdf
