Prediksi Keaslian Tanda Tangan Menggunakan Metode Prewitt dan Support Vector Machine Di Program Studi Teknik Informatika Institut Teknologi PLN
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
Ramadanty, Nur Anditha
Palupiningsih, Pritasari
Yosrita, Efy
Palupiningsih, Pritasari
Yosrita, Efy
Subject
Teknik Informatika
Datestamp
2022-09-27 02:45:52
Abstract :
Signature is evidence of an identity of a person whose individual signature is different and has distinct characteristics or characteristics. A college student at the institute technology PLN needs a lecturer signature for every semester To authenticate the exam plan card (crew). But autographs were done the 2019 by irresponsible students who put their own interests ahead of their own, in order to minimize the presence of autographs, it was necessary to create a system that could predict the authenticity of autographs. The study is conducted in several stages to predict the authenticity of the signature, the image preapprocess phase consists of changing the image of RGB to grayscale and binerization. In this study, the methods used The method prewitt used was prewitt, which detects minute signatures and extracted shapes to measure the value of eccentricity and metric that were used for calculating the method of supporting vector machine. For the authentic prediction of the signature, the method of support vector machine (SVM) is divided into two categories: the original signature and the false signature. The training data that is used as many as 10 signatures consisting of 5 original signatures and 5 false signatures from different persons, while for testing data uses 5 signatures. Testing results for accuracy using a confusion matrix of 80%.
Signature is evidence of an identity of a person whose individual signature is different and has distinct characteristics or characteristics. A college student at the institute technology PLN needs a lecturer signature for every semester To authenticate the exam plan card (crew). But autographs were done the 2019 by irresponsible students who put their own interests ahead of their own, in order to minimize the presence of autographs, it was necessary to create a system that could predict the authenticity of autographs. The study is conducted in several stages to predict the authenticity of the signature, the image preapprocess phase consists of changing the image of RGB to grayscale and binerization. In this study, the methods used The method prewitt used was prewitt, which detects minute signatures and extracted shapes to measure the value of eccentricity and metric that were used for calculating the method of supporting vector machine. For the authentic prediction of the signature, the method of support vector machine (SVM) is divided into two categories: the original signature and the false signature. The training data that is used as many as 10 signatures consisting of 5 original signatures and 5 false signatures from different persons, while for testing data uses 5 signatures. Testing results for accuracy using a confusion matrix of 80%.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
