A lightweight Flask web application that detects whether an ID image has been tampered with or altered, using image processing techniques like Structural Similarity Index (SSIM) and contour thresholding from OpenCV.
This project demonstrates how classical computer vision can effectively detect document manipulation.
How to Use the App
Follow these simple steps to check if an ID has been tampered with 👇
🪪 Step 1: Go to the Website
đź”— Open the web app in your browser: https://id-card-authenticity-web.onrender.com/
📤 Step 2: Upload the ID Images
Click the “Choose File” button.
Upload the image:
The suspected/tampered ID (the one you want to check).
🧠Step 3: Click “Check Authenticity”
Hit the Check or Detect button.
The system will:
Compare both images using Structural Similarity (SSIM).
Apply Contour Thresholding to find visual differences.
Highlight regions that may have been altered.
đź“Š Step 4: View the Results
Once processing is done (usually in a few seconds):
You’ll see the SSIM similarity score (from 0 to 1).
A result image will display with highlighted areas showing possible tampering.
Example:
Score 0.95+ → Likely Authentic ✅
Score 0.85–0.94 → Minor Differences