Sudira, Hanif (2017) Forensik Digital Deteksi Copy Move Pada Citra Menggunakan Modifikasi Expanding Block Algorithm. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
5113100184-Undergraduate_Theses.pdf - Published Version Download (4MB) | Preview |
Abstract
Copy-move adalah suatu bentuk tindakan kriminal pada perangkat digital dengan cara pemalsuan citra, dimana pengguna menyalin suatu bagian citra lalu menempelkannya di bagian lain pada citra yang sama. Biasanya teknik copy-move digunakan untuk menutupi objek dari suatu citra dengan cara menempelkan objek yang serupa di sekitarnya dengan maksud dan tujuan tertentu yang dapat merugikan. Oleh karena itu diperlukan metode yang dapat mendeteksi serangan copy-move secara cepat dan efisien, terlebih citra sering menjadi barang bukti dalam pernyelidikan tindakan kriminal.
Tugas Akhir ini mengimplementasikan dua cara untuk mencari dominant feature pada citra yaitu image gray dominant feature dan image color dominant feature. Tugas Akhir ini menggunakan metode Averaging Filter dan Median Filter sebagai penghilang noise pada citra yang akan dideteksi menggunakan metode expanding block algorithm. Tugas Akhir ini juga menggunakan data masukan citra yang diganggu oleh noise uniform distribution sebesar 0.5%.
Hasil uji coba parameter yang dilakukan didapatkan bahwa nilai parameter terbaik yang digunakan pada Tugas Akhir ini adalah numBuckets 12000, pVal 0.50, blockSize 16 dan minArea 90. Untuk hasil uji coba performa didapatkan bahwa nilai MSE terendah 1223.25, nilai similarity terbesar 99.27% dan nilai akurasi terbesar 93.06%. Penggunaan filter pada metode expanding block algorithm akan menurunkan performa serta metode expanding block algorithm dapat digunakan untuk mendeteksi citra gangguan noise uniform distribution 0.5% dengan menggunakan image color dominant feature.
==========================================================================================================================================
Copy-move forgery is a form of criminal acts digital device specific of image tampering where the part of the image is copied and pasted in another part of the same image. This technique is usually used to cover up an object in the image by attaching similar object around it. Therefore developing a method to verify and detect manipulated image became very important, especially because of the image as evidence in the investigation of criminal acts.
This undergraduate thesis implements two ways to search for a dominant feature in the image that is image gray dominant feature and image color dominant feature. This undergraduate thesis project using Averaging Filter and Median Filter as relieving noise at the image to be detected using the method of expanding the block algorithm. This undergraduate thesis also uses input data image plagued by noise uniform distribution of 0.5%
The trial results showed that the best parameters values used in this undergraduate thesis are numBuckets 12000, pVal 0.50, blocksize 16 and minArea 90. For the test results showed that the performance of the lowest MSE value 1223.25, the largest similarity value of 99.27% and value the accuracy of 93.06%. The use of filters on expanding block algorithm method will decrease performance. expanding block algorithm can be used to detect image noise with a uniform distribution of 0.5% by using image color dominant feature.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Uncontrolled Keywords: | Copy-move, filter, paramater, dominant feature, expanding block algorithm |
| Subjects: | T Technology > T Technology (General) T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TR Photography |
| Divisions: | Faculty of Information Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Hanif Hanif Sudira |
| Date Deposited: | 14 Mar 2017 02:35 |
| Last Modified: | 05 Mar 2019 04:04 |
| URI: | http://repository.its.ac.id/id/eprint/1830 |
Actions (login required)
 |
View Item |