Pengenalan Pola Menggunakan Metode Cellular Automata (Studi Kasus : Verifikasi Tandatangan) Pattern Recognition Using Cellular Automata (Signature Verification) Diajukan Untuk Memenuhi Salah Satu Syarat Memperoleh D

adiananda, Retantyo Wardoyo

2016 | Disertasi | FMIPA

a pattern recognition technique in the field of machine learning and can be defined as "the act of taking raw data and act upon data classification". Many research has been done on the topic of pattern recognition to verify the signature using a variety of methods one of which is by using artificial neural networks. This research use cellular automata method for finding and extracting the feature of a signature image. Implementation of this research will be realized in the form of a pattern recognition software that can be used to verify the authenticity of the signature. This research will use the data signature in the form of a signature image files. To determine the characteristics of each image vector signatures, will be used the method of measuring the frequency of occurrence of the patterns that is owned by the neighboring cellular automata. Neighbor pattern used in this study is Neumann 5 neighbors model. In this study used the data 9 units signatures of 57 respondents. Six signatures used for calculating the threshold value (NA). Tests on the proposed method is done by using three pieces of original signatures of each of the respondents and 10 signatures were taken randomly from the respondents were registered. From the results obtained the results of tests performed by 88.30% precision, recall amounted to 65.36% and the level of system accuracy is 71.31%.

Kata Kunci : pattern recognition, cellular automata, verification, signature