A Robust Approach to Authentication of Handwritten Signature Using Voting Classifier
Kamlesh Kumari, Sanjeev Kumar Rana · Journal of Computational and Theoretical Nanoscience · 2020
This research paper’s main motive is to improve the recognition rate of Offline Signature Verification system. In our research, decisions of three classifiers i.e., Multilayer Perceptron, Random Forest Classifier and Naive Bayes are combined using voting classifier to determine the output. The softwares used for this research are WEKA and Matlab. Performance of this approach is tested on CEDAR dataset for writer dependent model. Overall recognition rate of whole dataset of 55 users is 91.25%. Out of the dataset, the recognition rate of 45 users is above 85%.