PENGEMBANGAN PRINTER FORENSIK UNTUK IDENTIFIKASI DATA DOKUMEN CETAK

Kurniati, Florentina Tatrin (2015) PENGEMBANGAN PRINTER FORENSIK UNTUK IDENTIFIKASI DATA DOKUMEN CETAK. S2 thesis, UAJY.

[img] Text (Halaman Judul)
MTF002007.pdf

Download (1MB)
[img] Text (Bab I)
MTF102007.pdf

Download (306kB)
[img] Text (Bab II)
MTF202007.pdf

Download (289kB)
[img] Text (Bab III)
MTF302007.pdf

Download (1MB)
[img] Text (Bab IV)
MTF402007.pdf
Restricted to Registered users only

Download (1MB)
[img] Text (Bab V)
MTF602007.pdf
Restricted to Registered users only

Download (286kB)
[img] Text (Bab VI)
MTF502007.pdf

Download (19MB)

Abstract

In the process of identification documents that are evidence of the particular case, by analyzing the characteristics contained in the document. The process of getting characterize the evidence carried out by extracting documents, previously digitally converted to forms through the process of scanning. Characteristics contained in the documents of evidence when compared with the characteristics of the comparator (unpredictable), and both expressed the same then the results of the analysis is the printer used to print the document came from the same printer. However, the identification process is also constrained when unexpected printer has indicated the type and the same type. For that we need for the development of forensic printer to identify the problems. In this study developed a printer forensic identification process, especially compared with the comparative evidence documents originating from the printer that has the same type and type. Tests carried out using five types of printer, one of which is a printer used to print the proof. The data used for comparison are each printer will print the data taken as much as 6 sheets with the contents of the document that has been determined, and the data is duplicated into digital through the scan with the quality of 600 dpi, 600 dpi aims to make use of the duplication process as much as possible approached document original. From the documents of each sheet comparators used traits as much as 9 different characters as well as the evidence used to document as much as 9 different characters with 6 sampeli. So the whole character is used as a benchmark of 270 characters, and evidence of as many as 54 data. The next process the document is processed using Canny edge detection method to highlight the characteristics before extracted using GLCM., This extraction results obtained characterize as many as 162 proof characteristics and thus also to the printer from one to five. Based on testing performed for the results obtained with FCM centroid values between evidence and comparative approach the same value as compared to other comparable printers. Additionally seen that kind of character "H" and "o" can not recognize. As for the character "k", "b" and "image" can be used for comparison with the good. EER value of 0.4 on test results obtained with average accuracy rate 80.0707%.

Item Type: Thesis (S2)
Uncontrolled Keywords: Forensic printer, Edge Detection, GLCM, FCM
Subjects: Magister Teknik Informatika > Soft Computing
Divisions: Pasca Sarjana > Magister Teknik Informatika
Depositing User: Editor UAJY
Date Deposited: 29 Jun 2015 09:57
Last Modified: 29 Jun 2015 09:57
URI: http://e-journal.uajy.ac.id/id/eprint/7549

Actions (login required)

View Item View Item