- Notifications
You must be signed in to change notification settings - Fork0
eyuuab/Frame_level_video_forgery_detaction
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
This project analyzes video frame differences to detect potential tampering, such as duplicate or frozen frames. Using OpenCV and NumPy, the script compares a tampered video with a real video and visualizes the results.
The script usesframe difference analysis:
- Convert each video frame to grayscale to simplify processing.
- Calculate the absolute difference between consecutive frames.
- Count the number of non-zero pixels in the difference to measure changes.
- Flag frames with minimal changes (below a threshold) as potential duplicates or frozen frames.
This approach highlights anomalies that may indicate video tampering.
- Python 3.x
- Required Libraries:
- OpenCV (
cv2) - NumPy (
numpy) - Matplotlib (
matplotlib)
- OpenCV (
Install dependencies using:
pip install opencv-python-headless numpy matplotlib