Recognition of person’s character through the shape of signature using Radial Basis Function Neural Network (RBFNN) method
Faisol Faisol, Ahmad, Halumatus Sakdiyah, Qurratul Aini, Kuzairi Kuzairi, Tony Yulianto · Journal of Physics Conference Series · 2020
Abstract Signatures are a marker or identity that exists in a document. Signatures have an important role in verifying and legalizing documents. The purpose of this study is to apply image processing techniques to signatures and to identify patterns of signature images based on entropy values. The stages of the research include taking respondents’ data in the form of analog image signatures, then the acquisition of digital signature images by scanning the signatures. The next step is to convert digital hand images from true color to binary. The last step is calculating the entropy value, recording the entropy value calculation time using Matlab software and looking at the distribution of entropy values from each signature image. From this method, the result of 50 data of signatures taken from students of the Islamic University of Madura (UIM) of the FMIPA, there are 11 to 1st type, 1 to 2nd type, 13 to 3th type, 4 to 4th type, 9 to 5th type, 3 to 8th type, 7 to 9th type and 2 to 11th type.