AI — AI-driven Authentication and Feature Education
Sri Lankan Currency Note Recognition System
A Flask-based application that authenticates Sri Lankan currency using digital image processing and machine learning. The system verifies authenticity and explains security elements using CV and deep learning, tailored to local notes and their cultural markers.
- Role
- Developer
- Client
- Academic Project
- Year
- Nov 2024
- Scope
- 1 month
- Team
- Solo
- Tools
- Machine Learning · Deep Learning · Flask · Python · OpenCV · TensorFlow · PyTorch
Outcome
- Implemented robust counterfeit detection system
- Created educational interface for security features
- Added advanced image processing capabilities
- Integrated YOLO for symbol detection and OCR
Tags
What I did
- 01Built Flask backend integrating ML inference endpoints
- 02Implemented OpenCV preprocessing and feature extraction
- 03Trained/Integrated YOLO for symbol detection on notes
- 04Added OCR and descriptive UI for security elements
Key features
- 01Real-time authentication feedback
- 02Security feature detection (watermarks, UV, micro-text)
- 03Educational descriptions for cultural/security markers
- 04Detection and classification with YOLO and OCR