Copy-Move Counterfeit Detection: Duplication Method

Copy-Move falsify detection using the Duplication Method in grayscale by applying the Principal Component Analysis in python3

This project was an adaptation of the work by Rahmat Nazali (github:rahmatnazali), "Copy-Move Detection on Digital Image using Python", with the objective of carrying out detections of the Copy-Move digital image forging method using only a gray scale of the images via the Duplication Method, using a principal component through Principal Component Analysis in python3; using the results to compare and study about both codes. His work can be found at his github rahmatnazali/image-copy-move-detection.

The databases used were taken from the github above and at the following link CIVIP Group - Copy-Move Forgery Dataset.

All credits granted to Rahmat Nazali.

@TalesNogueira

Execution of the algorithm

To run the algorithm, just follow the steps below. Remembering that the folder containing the PNG datasets to be checked is "..\Copy_Move_Counterfeit_Detection\dataset\multi_paste".

1. Install python3+

2. Install libraries listed in "library.txt"

3. Execute "main.py"

4. Choose an image by its index number listed

5. Choose a number positive and non-zero to be the blocksize

6. Enjoy yours results in the "output" folder