AI-Powered Signature Verification System Using Deep Learning

Overview

The Signature Verification System is an advanced web application designed to verify the authenticity of handwritten signatures using deep learning technologies. Built with Python and powered by a Convolutional Neural Network (CNN), this system offers both image upload and live webcam signature capture features.

Key Features

• Drag-and-Drop Image Upload: Easily upload signature images via intuitive drag-and-drop interface or traditional file selection.
• Real-Time Webcam Capture: Capture live signatures directly from your webcam for instant verification.
• Deep Learning Based Verification: Uses a pre-trained CNN model to analyze and determine if a signature is genuine or forged.
• Confidence Score Display: Each verification result includes a confidence percentage to indicate accuracy.
• Modern & Responsive UI: Clean, mobile-friendly design ensures seamless user experience across all devices.
• Error Handling: Comprehensive handling for file uploads, model predictions, and device access issues.

Applications

• Banking Sector: Verify digital signatures during online transactions or document approvals.
• Legal Documentation: Authenticate electronic signatures in legal contracts and agreements.
• E-Governance: Securely validate citizen identities in government portals.
• E-Signature Platforms: Enhance trust and security in digital signing services.

Technology Stack

• Backend: Flask (Python)
• Frontend: HTML5, CSS3, JavaScript
• Machine Learning: TensorFlow/Keras CNN Model
• Model File: signature_cnn_model.h5
• Image Processing: OpenCV

Installation Guide

1. Clone the repository: Extract the provided zip file
2. Navigate into the directory: cd signature-verification-system
3. Create virtual environment: python -m venv venv
4. Activate the environment:
   • Linux/macOS: source venv/bin/activate
   • Windows: venv\Scripts\activate
5. Install dependencies: pip install -r requirements.txt
6. Ensure the trained model file signature_cnn_model.h5 exists in the models folder.

How to Use

1. Run the Flask app: python app.py
2. Open browser and go to http://localhost:5000
3. Upload an image or use your webcam to capture a signature
4. Click “Verify Signature” and view detailed results including authenticity and confidence score

Security Measures

• Secure filename sanitization
• File type and size restrictions
• Error message sanitization
• No sensitive data logged

Contributing & Licensing

This project is open-source under the MIT License. Contributions are welcome via Pull Requests. Please ensure code adheres to the existing structure and standards.

Whether you're building a secure e-signature platform or verifying digital documents, this AI-powered signature verification system provides robust and scalable solutions for modern businesses and institutions.