EKSTRAKSI FITUR PADA CITRA TANDA TANGAN SEBAGAI IDENTITAS PEMILIKNYA MENGGUNAKAN LOCAL RANGE FILTER (LRF) --

ABSTRACT
A signature is a sign to represent the name written with the author's own
signature as a personal marker. Usually a signature is used to approve an
agreement on a document sheet as proof of the identity and willingness of the
owner of the signature. In general, the affixing of signatures is carried out in state
administrative activities such as cash withdrawal transactions, deposits, clearing
demand deposits, and so on. To analyze the signature image, an image pattern is
used which is built through a feature analysis process. This signature image
research was conducted using the Local Range Filter (LRF) feature extraction
method. Performance test is used to measure the success of the research method.
The results showed that the test using 25 signature data resulted in the percentage
of K-Nearest Neighbor (KNN) classification of 68%.
Keywords: Feature extraction, signature image, Local Range Filter (LRF), KNearest
Neighbor (KNN)