Implementation of Inner Product to Analyze Digital Handwriting based on Texture Traits
R. Davin RP., Dian Pratiwi, Syaifudin, R. Trubus, D. L. P. Rizky · 2017
Nowadays, crimes against the falsification of document is keep growing. For example, falsification of signature on an important certificate, such as, land certificate or educational certificate. For that reason, implementation of inner product to analyze digital handwriting based on texture traits is the one of important things to do to against the falsification of document or certificate. This study was conducted with the purpose of building a web application system that can know the owner of the document through his handwriting form, in which case, it is expected that falsification of documents can be prevented. The method that applied in this study consisted of writing stage analog to digital conversion, manual segmented into the size of 500x200 pixels every data, feature extraction stage, and the percentage calculation of similarity through the similarity measures with inner product method. This study has been conducted on 100 data of 20 correspondent, which each correspondent gives 5 handwriting data. Furthermore, the data will be divided by the proportions of 3:2 for the reference and the tester. Then, the tester data will be tested with the reference data, and take three ranks of the closest results. From 40 data, 22 data identified the owner managed appropriately. Which mean, the percentage of accuracy is 55%.