Analisis perbandingan metode discrete fourier transform (dft) dan discrete radon transform (drt) pada identifikasi pola tanda tangan

Wina, Fitriani (2018) Analisis perbandingan metode discrete fourier transform (dft) dan discrete radon transform (drt) pada identifikasi pola tanda tangan. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

Preview	Text Abstract.pdf - Accepted Version Download (274kB) | Preview
Preview	Text Abstrak.pdf - Accepted Version Download (281kB) | Preview
	Text Cover.pdf - Accepted Version Download (867kB)
Preview	Text BAB I.pdf - Accepted Version Download (292kB) | Preview
Preview	Text BAB II.pdf - Accepted Version Download (531kB) | Preview
Preview	Text BAB III.pdf - Accepted Version Download (232kB) | Preview
	Text BAB IV.pdf - Accepted Version Restricted to Registered users only Download (969kB)
Preview	Text BAB V.pdf - Accepted Version Download (181kB) | Preview
Preview	Text Daftar Pustaka.pdf - Accepted Version Download (307kB) | Preview

Abstract

Knowing importance of signature in national administration activity as identity evidence and willingness from the owner of its signature, so identifications are needed to recognize and differentiate from each different person based on characteristics from its signature. With current technology development, identification of an signature patterns can be done with computer helps. But computer not directly perform the identification process, it’s require recognition patterns that can be done with extracting signatures features. One of the features that can be extracted from an signature is the result of a transformation. DFT and DRT are some extraction methods that used in identification of signature patterns. DFT method has advantage when image result of the transformation being inverted become original image will remain the same. Whereas the advantage of DFT method is enable to calculate only short interval in Θ. However both method laxity is has long computation time. Value that used for classification in each method is mean and standard deviation. Whereas MLP is classification that used in both method. The application of DFT and DRT in data resulted low level of accuracy, DFT is 9.5556% while DRT is 39.5556%. %. It can be affected by non unique value from each class, so the difficulty test's data to determine the real class. The amount of data quantity and number of classes can also affect the accuracy. Keywords: DFT, DRT, Identification, Pattern, Signature

Actions (login required)

View Item