/ForgedVideoDetector

A tool to detect whether your video has been forged (tampered with).

Primary LanguageMATLAB

VIDEO-FORGERY-DETECTION


			READ ME FOR RUNNING OF VIDEO FORGERY DETECTION MATLAB CODE

  1. Install Matlab R2012a or above on your PC running any of the standard Operating System(Windows 7,8,8.1,10).

  2. Include the 'jpeg_read' library in your matlab to run the files included in the package.

  3. Run the 'feature_original.m' file to extract and store the 'ccpev548' extracted features in a 'cover.mat' file of the non-forged videos.

  4. Run the 'feature_forged.m' file to extract and store the 'ccpev548' extracted features in a 'stego.mat' file of the forged videos.

  5. Run the 'trainer.m' file to train the system on your training set which creates a 'trained_ensemble.mat' file, which stores the data in this variable to be used further in the testing purposes.

  6. Finally, run the 'tester.m' file to test on a particular video by giving input the feature extracted matrix (.mat) file for that video and also the 'trained_ensemble' variable to give as input, formed in the previous file (trainer.m).