IDENTIFIKASI CITRA TANDA TANGAN ASLI DAN PALSU
MENGGUNAKAN GRID ENTROPY, PRINCIPAL COMPONENT
ANALYSIS, DAN SUPPORT VECTOR MACHINE
Total View This Week0
Total View This Week0
Institusion
Universitas Pembangunan Nasional Veteran Jakarta
Author
Riduwan Purnaminyan, .
Subject
QA75 Electronic computers. Computer science
Datestamp
2021-12-21 07:36:16
Abstract :
Signature is a person's writing that has a certain writing style, is symbolic, and tends to be different for everyone. Signatures can be used on a document as evidence that someone agrees to the contents of the document and is bound by the rules and responsibilities therein. By knowing how much influence the signature has, it often makes someone want to imitate or fake a signature for personal gain. Therefore, it is necessary to have a system that can help solve this problem in identifying the authenticity of the signature. In this study, grid entropy feature extraction method is used to extract signature image characteristic features, principal component analysis dimension reduction is used to reduce computational load without significantly reducing accuracy, and lastly support vector machine classification algorithm is used to produce optimal classification results on small scale datasets. The results obtained are 95% accuracy, 96.7% precision, and 94.7% f1-score so that the model can be said to be successful in identifying the original signature image and the fake signature image well.
Signature is a person's writing that has a certain writing style, is symbolic, and tends to be different for everyone. Signatures can be used on a document as evidence that someone agrees to the contents of the document and is bound by the rules and responsibilities therein. By knowing how much influence the signature has, it often makes someone want to imitate or fake a signature for personal gain. Therefore, it is necessary to have a system that can help solve this problem in identifying the authenticity of the signature. In this study, grid entropy feature extraction method is used to extract signature image characteristic features, principal component analysis dimension reduction is used to reduce computational load without significantly reducing accuracy, and lastly support vector machine classification algorithm is used to produce optimal classification results on small scale datasets. The results obtained are 95% accuracy, 96.7% precision, and 94.7% f1-score so that the model can be said to be successful in identifying the original signature image and the fake signature image well.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
