Machine Learning · 2024
ForgeryGuard
An image authenticity verification system using machine learning — detecting forgeries through Error Level Analysis and neural classification.
Overview
ForgeryGuard is a web-based forensics tool that analyzes uploaded images for signs of digital manipulation. It combines classical Error Level Analysis (ELA) with a convolutional neural network to classify images as authentic or forged, presenting confidence scores and visual heatmaps to the user.
- Error Level Analysis to detect inconsistent JPEG compression artifacts
- CNN classifier trained on the CASIA2 image forgery dataset
- Visual heatmap overlay highlighting suspected manipulated regions
- Flask web interface — drag-and-drop upload, instant result display
Stack
Python
Flask
TensorFlow / Keras
OpenCV
HTML / CSS / JS
NumPy · Pillow
Navigation